Skip to main content
NCBI home page
Neuro-Oncology logoLink to Neuro-Oncology
. 2012 Jun;14(Suppl 1):i26–i32. doi: 10.1093/neuonc/nos098

DIFFUSE INTRINSIC PONTINE GLIOMA (DIPG)

PMCID: PMC3483344
Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-01. HYPOFRACTIONATED RADIOTHERAPY FOR PEDIATRIC DIFFUSE INTRINSIC PONTINE GLIOMA (DIPG): A PROSPECTIVE CONTROLLED RANDOMIZED TRIAL

Mohamed Zaghloul 1, Soha Ahmed 1, Eman Eldebaway 1, Amr Mousa 1, Amr Amin 1, Nada Elkhateeb 1, Mohamed Sabry 1

Abstract

BACKGROUND: Pediatric Diffuse intrinsic pediatric glioma (DIPG) remained dismal regardless of the new therapeutic and technical attempts. PURPOSE: To investigate the efficacy and toxicity of hypofractionated radiotherapy in pediatric DIPG compared to conventional radiotherapy and to determine the prognostic factors for its overall (OS) and progression-free survival (PFS). PATIENTS AND METHODS: Sixty four children, below the age of 18 years, who presented to Children's Cancer Hospital, Egypt (CCHE) during the period July 2007 and July 2011 were randomized into 2 groups. The hypofractionated group received 39 Gy in 13 fractions in 21/2 weeks and the conventional arm receiving 55.8 Gy in 31 fractions in 6 weeks using conformal radiotherapy technique. The two arms were not different in age, gender, performance status and tumor volume. RESULTS: Thirty two children were randomized in each arm. The median OS for the whole group was 9.5 ± 1.0 months: 7.4 ± 1.0 months for the hypofractionated arm and 9.9 ± 1.0 months for the conventional arm. The whole group has median PFS of 7.3 ± 0.8 months; 7.0 ± 1.4 months for the hypofractionated and 7.7 ± 1.1 for the conventional arm. On the other hand, the 1-year and 2-year OS were 41.4 ± 9.2% and 28.4 ± 8,8% in the hypofractionated arm and 36.2 ± 8.7% and 32.3 ± 7.8% in the conventional arm. None of these differences was statistically significant. Furthermore, none of the tested factors (age, gender, performance status, tumor volume, radiation volume or tumor extension) proved to have statistically significant influence on OS or PFS. All patients showed marked degree of improvement in symptoms and signs with earlier response in the hypofractionated arm. The immediate and delayed side effects were not different between the 2 arms. CONCLUSIONS: Hypofractionated radiotherapy had similar overall, progression-free survival and delayed effect (in the long survivors) to conventional fractionation. It offers less burden on the patients, their families and the treating machines and departments.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-02. THE EFFICACY OF THE BIOPSY OF INTRINSIC BRAINSTEM LESIONS FOR DECISION-MAKING OF THE TREATMENTS

Hideki Ogiwara 1, Nobuhito Morota 1

Abstract

OBJECT: The management of intrinsic brainstem lesions remains controversial. Some considered a biopsy of intrinsic brainstem lesions not safe and magnetic resonance imaging (MRI) is sufficient to diagnose diffuse brainstem glioma and make treatment decisions. However, the accuracy of MR imaging-based diagnosis of diffuse brainstem gliomas has not been fully confirmed yet. The aim of this study was to review efficacy of biopsy of intrinsic brainstem lesions for decision-making of the treatments. METHOD:. We retrospectively analyzed surgical results of pediatric patients with intrinsic brainstem lesions who underwent biopsy at the Division of Neurosurgery, National Center for Child Health and Development, Tokyo, over the period of March 2008 to November 2011. RESULTS: Seven patients underwent biopsy for intrinsic brainstem lesion. Preoperative MR revealed focal enhancing lesion in 2 patients, diffuse enhancing lesion in 3, and diffuse non-enhancing lesion in 2. Biopsy was performed through midline suboccipital craniotomy in 6 patients and through retrosigmoid craniotomy in one, under the cortico-bulbar tract motor evoked potential (CBT-MEP) monitoring. No intra-operative or postoperative complications were observed. Histological evaluation revealed diffuse brainstem glioma in 5 patients, PNET in 1, multiple screlosis (MS) in 1. For a patient with PNET chemotherapy and radiotherapy were effective and the patient has been stable for 9 months without recurrence. For a patient with MS steroid therapy was effective. CONCLUSIONS: The biopsy of intrinsic brainstem lesions using craniotomy is considered to be safe and effective for selecting appropriate adjuvant therapy.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-03. DIFFUSE PONTINE PNET: CONFOUNDING THE DIAGNOSIS OF DIPGS

Alexandra Sufit 1, Andrew Donson 1, Diane Birks 1, Purvi Patel 1, Nicholas Foreman 1, Michael Handler 1

Abstract

INTRODUCTION: Diagnosis of typical diffuse pontine tumors, as a standard, is by scan. This cannot distinguish primitive neuroectodermal tumors (PNETs) of the pons from diffuse intrinsic pontine gliomas (DIPGs). We present two cases of PNETs diffusely in the pons and show molecular evidence these are indeed PNETs and distinct from biopsied atypical DIPGs. METHODS: Pediatric brain tumor samples were collected and snap frozen in the operating room. All were analyzed histologically, and for cluster analysis and aberrant gene expression via Ambion Microchip HG-U133plus2. Biopsied diffuse tumors of the pons were selected, comparing them against other intracranial tumors. RESULTS: The histology of 5 atypical diffuse tumors of the pons was consistent with malignant gliomas, and in 2 young children (<3 years) was consistent with PNET. Clustering analysis demonstrated that pontine PNETs were most closely related to PNETs of the supratentorial region, not clustering with the DIPGs in the study. Over 200 genes, many of which are involved in transcriptional regulation, were uniquely up-regulated in brainstem PNETs compared to 10 other types of pediatric brain tumors, including DIPGs and other PNETs, at an FDR statistical significance of < 0.01. CONCLUSION: As shown by the cluster analysis and the individual gene expression analysis, pontine PNETs are distinct from DIPGs. This suggests that these tumors should be treated differently than DIPGs. Since diagnosis by scans is not reliable and the biology of the tumors is disparate, biopsies should be performed to enable proper diagnoses and more effective treatments.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-04. NIMOTUZUMAB AND VINORELBINE CONCOMITANTLY TO RADIATION AND AS MAINTENANCE FOR DIFFUSE PONTINE GLIOMA IN CHILDHOOD: PROMISING RESULTS ON A SERIES OF 20 PATIENTS

Maura Massimino 1, Veronica Biassoni 1, Lorenza Gandola 1, Elisabetta Schiavello 1, Emilia Pecori 1, Paolo Potepan 1, Ferdinand Bach 2

Abstract

BACKGROUND: After a previous trial with nimotuzumab/radiation and continuing for a total of 37 patients from 2006 to 2009 according to this new combination, we obtained a median PFS of 7 months and a median OS of 11 months, i.e. consistent with best literature data and those reported previously by ourselves. All treatment was given on an outpatient basis without any side effects correlated with the use of nimotuzumab. Two/37 children were alive without tumor progression 31/39 months after their diagnosis. In a subsequent mono-institutional experience, with a view to exploring add-on strategies, we began a pilot protocol on a compassionate case-by-case basis using nimotuzumab with vinorelbine, combined with radiation, and adopting consolidation courses with the same timing as in the previous international protocol METHODS: Vinorelbine was adopted at a dose of 20 mg/sqm/weekly together with nimotuzumab at the standard dose of 150 mg/sqm during the 6 weeks when radiotherapy was delivered, and 25 mg/sqm in any other week, with the same dose of nimotuzumab during the consolidation courses, planned until tumor progression or for a total of two years. RESULTS: From August 2009, we have treated 20 children, 12 males and 8 females, age range of 2-17 years (median 7), enrolled according to the standard MRI inclusion criteria. After a median follow-up of 14 months, 12/20 were alive, their PFS/OS at 12 months were 32%/81%, respectively; OS at two years was 28%. According to MRI, 12 had PR, 5 SD with 100% symptom amelioration, 1 early PD and 2 local PR but spinal dissemination. The nimotuzumab/vinorelbine combination was very well tolerated, with no acute side-effects. All children were treated on an outpatient basis. CONCLUSIONS: This combination has promise, with significant differences with previous institutional and literature reported experiences.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-05. HYPOFRACTIONATION VERSUS CONVENTIONAL RADIOTHERAPY FOR NEWLY DIAGNOSED DIFFUSE INTRINSIC PONTINE GLIOMA: A MATCHED COHORT ANALYSIS

Geert O Janssens 1, Marc H Jansen 2, Selmer J Lauwers 1, Peter J Nowak 3, Foppe R Oldenburger 4, Eric Bouffet 5, Frank Saran 6, Karin Kamphuis van Ulzen 7, Erik J van Lindert 8, Jolanda H Schieving 9, Tom Boterberg 10, Gertjan J Kaspers 2, Paul N Span 1, Johannes H Kaanders 1, Corrie E Gidding 11, Darren Hargrave 12

Abstract

BACKGROUND: Despite conventional radiotherapy, 54 Gy in single doses of 1.8 Gy (54/1.8 Gy) over 6 weeks, most children with diffuse intrinsic pontine glioma (DIPG) will die within one year after diagnosis. To reduce patient burden, we investigated the role of hypofractionation radiotherapy, given over 3-4 weeks. A 1:1 matched-cohort analysis with conventional radiotherapy is performed to assess response and survival. METHODS: Twenty-seven children, aged 3-14, were treated according to one of 2 hypofractionation regimens over 3-4 weeks (39/3 Gy, N = 16 or 44.8/2.8 Gy, N = 11). All patients had symptoms for ≤3 months, ≥2 signs of the neurological triad (cranial nerve deficit, ataxia, long tract signs) and characteristic features of DIPG on MRI. Twenty-seven patients fulfilling the same diagnostic criteria and receiving at least 50/1.8-2.0 Gy were eligible for the matched-cohort analysis. RESULTS: With hypofractionation radiotherapy, the overall survival at 6, 9 and 12 months was 74%, 44% and 22%. Progression-free survival at 3, 6 and 9 months was 77%, 43% and 12%. Temporary discontinuation of steroids was observed in 21/27 (78%) patients. No significant difference in median overall survival (9.0 vs. 9.4 months; P = .84) and time to progression (5.0 vs. 7.6 months; P = .24) was observed between hypofractionation vs. conventional radiotherapy, respectively. CONCLUSIONS: For patients with a newly diagnosed DIPG, a hypofractionation regimen, given over 3-4 weeks, offers equal overall survival with less treatment burden compared to a conventional regimen of 6 weeks.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-06. DIFFUSE INTRINSIC PONTINE GLIOMA TREATED WITH PROLONGED TEMOZOLOMIDE AND RADIOTHERAPY – RESULTS OF THE UNITED KINGDOM TRIAL (CNS 2007 04)

Simon Bailey 1, Andrew Howman 2, Barry Pizer 3, Dee Harris 2, Deborah Jones 2, Pamela Kearns 2, Susan Picton 4, Frank Saran 5, Keith Wheatley 2, Michael Gibson 2, Adam Glaser 4, Daniel Connolly 6, Darren Hargrave 7

Abstract

Diffuse intrinsic pontine glioma (DIPG) has a dismal prognosis with most children dying within a year of diagnosis. No chemotherapy regimen has so far resulted in any significant improvement over standard radiotherapy. Temozolomide, has shown efficacy in non-brainstem high grade glioma, but it is known that O-6-methylguanine-DNA methyltransferase (MGMT) can mediate resistance. In this protocol a prolonged low dose regimen (21/28 day) of temozolomide was studied with the aim of better depleting MGMT levels and increasing cytotoxicity. Forty-three patients with a defined clinico-radiological diagnosis of DIPG were treated with radiotherapy and concomitant temozolomide (75 mg/m2) after which up to 12 courses of 21 days of adjuvant temozolomide (75 -100 mg/m2) were given 4 weekly. The trial was a 2-stage design which passed interim analysis. Concomitant Quality of Life assessment was mandated as part of the study. Median age at diagnosis was 8 years (2-20). 81% of patients presented with cranial nerve abnormalities, 76% with ataxia and 57 % with long tract signs. Mean Karnofsky/Lansky score was 79 (10-100). Patients received a median of 3 courses of maintenance temozolomide, 5 received all 12 courses and 7 did not start maintenance. Three patients were withdrawn from the study due to hematological toxicity and 10 had a dose reduction on at least one course. No other significant toxicity related to temozolomide was noted. Overall survival (95%CI) was 0.56 (0.40, 0.69) at 9 months, 0.35 (0.21, 0.49) at 1 year and 0.17 (0.07, 0.30) at 2 years. Median survival was 9.5 months (7.5, 11.4). Temozolomide, even when administered in this alternative regimen, offers no benefit to standard radiotherapy in DIPG. Recent reports using either biopsy or autopsy tumour material are starting to elucidate the underlying molecular biology of DIPG and future trials will aim to select and target novel therapies based on this emerging data.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-07. THE COMBINED THERAPY OF BRAIN STEM GLIOMA 6 CASES UNDER 15 YEAR-OLD IN HYOGO PREFECTURAL KOBE CHILDREN'S HOSPITAL

Atsufumi Kawamura 1, Tatsuya Nagashima 1, Kazuki Yamamoto 1, Junichi Sakata 1

Abstract

INTRODUCTION: Brain stem glioma is highly malignant pediatric brain tumors with poor prognosis and only radiation therapy has been approved. This report reflects our experience at the Hyogo Prefectural Kobe Children's Hospital (Japan) with temozolomide (TMZ) with local radiation therapy of six brain stem glioma from 2010 to 2012. METHODS: Children aged under 15 years with newly diagnosed brain stem glioma were treated with standard radiotherapy and TMZ at 75 mg/m2/day for 6 weeks. After completion of the radiotherapy, TMZ (150-200 mg/m2) was administered once per day for five consecutive days in each 28-day cycle. The median age was seven years (range, 6-10 years); four patients were male and two were female. The mean follow-up period was 7 months (range, 3-10 months). All diagnoses were established using neuroradiological studies. RESULTS: Five patients showed a partial response (PR), whereas one patient exhibited stable disease (SD). The median progression-free survival (PFS) was 5.6 months. Five of 6 cases showed the great improvement in Karnofsky's performance status and returned to school life. In one case, tumor indicated regrowth in 7 months after irradiation but the progression had been stagnant for 5 months with increase of TMZ (200 mg/m2). The complication was only constipation. CONCLUSION: In order to improve outcome and preserve quality of life of poor prognosis cases less than 12 months, temozolomide should be reconfirmed the role in treatment of pediatric brain stem glioma.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-08. CLUES TO PONTINE GLIOMAGENESIS FROM CHILDHOOD PONTINE GLIOGENESIS

Robert Lober 1, Morgan Freret 2, Paul Fisher 1, Michael Edwards 1, Kristin Yeom 1, Michelle Monje 1

Abstract

The region- and age-specific nature of diffuse intrinsic pontine glioma (DIPG) suggests that the underlying pathophysiology involves dysregulation of a postnatal neurodevelopmental process. We have recently identified a novel glial precursor cell population in the ventral pons of both mice and humans, present only during infancy and middle childhood. This cell is detected between six and nine years of age, corresponding to the peak age of brainstem gliomas formation. Using genetic mouse models, we determined that this precursor cell type is a glial progenitor that contributes to a second, previously unrecognized, wave of mid-childhood myelination of the corticospinal tracts under the control of the Hedgehog signaling pathway. Using diffusion tensor imaging, we evaluated white matter tract development in the human pons throughout childhood in 74 children ranging in age from one month to 18 years of age. All subjects included had normal neuroimaging and no neurological symptoms or signs. In all white matter tracts examined (corticospinal, transverse pontine fibers, medial lemniscus and superior cerebellar peduncle), there was an increase in fractional anisotropy between birth and two years of age, corresponding to the major wave of early postnatal myelination. In the corticospinal tracts only, there was also a second increase in fractional anisotropy between ages six and nine years, suggesting a second, previously unrecognized wave of human corticospinal tract myelination. We hypothesize that the spatiotemporal pattern of mid-childhood corticospinal tract myelination reflects the activity of Hedgehog-regulated glial progenitor cells susceptible to malignant transformation and development of brainstem gliomas.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-09. A COMBINED CLINICAL AND RADIOLOGICAL SURVIVAL PREDICTION TOOL FOR PONTINE GLIOMA

Marc Jansen 1, Esther Sanchez Aliaga 1, Erica Van Der Hoeven 1, Dannis Van Vuurden 1, Martijn Heymans 1, Corrie Gidding 2, Eveline De Bont 3, Roel Reddingius 4, Cacha Peeters-Scholte 5, Antoinette Schouten van Meeteren 6, Rob Gooskens 7, Bernd Granzen 8, Gabriel Paardekoper 9, Geert Janssens 2, David Noske 1, Frederik Barkhof 1, W Peter Vandertop 1, Gertjan Kaspers 1

Abstract

INTRODUCTION: Pontine gliomas (PG) have a dismal prognosis. Several (non-randomized) trials did not improve the dismal prognosis, although differences in overall survival have been reported. To determine whether differences in OS are due to real treatment effects or to distinct natural disease courses, a survival prediction tool is needed. METHOD: A nation-wide cohort study was applied to collect clinical and radiological data of patients aged 0-18 years with PG, confirmed by centrally reviewed MRI, excluding dorsally exophytic tumors, diagnosed in the Netherlands from 1990 until 2010. A Cox proportional-hazard model with backward regression was used to select prognostic factors to predict 12-months risk of death. The model performance was evaluated by using the area under the curve (AUC) and validated by bootstrapping (n = 250). Risk score (RS) categories with corresponding sensitivity (SE) and specificity (SP) were calculated. RESULTS: 121 patients with pontine glioma (PG) were eligible for analyses. Five predictors out of 16 variables were selected by backward regression (p < 0.15) and included into the model. The AUC was 0.67. The contribution to the RS to predict 12-months risk of death was: “symptom duration before diagnosis” (RS for every month = -1), “tumor homogeneity” (=-4), “percentage of pontine tumor involvement” (50-67% = +8; >67%= +12), “degree of basilar artery encasement” (180°-360° = -2; 360° = +5) and presence of ring-enhancement after gadolinium administration (+5). RS categories with corresponding sensitivity and specificity for the 12-months risk of death were: RS ≥17 (SP 100%, SE 14%); RS ≥11 (SP 88%, SE 57%); RS ≥5: (SE 91% SP 63%); RS ≥-1 (SE 95%, SP 38%); RS ≥ -7 (SE 100%, SP 0%). CONCLUSION: We have constructed a clinically relevant tool to predict risk of death in pontine glioma patients at diagnosis. This model needs to be validated in new patients, and once validated can be used to compare populations within and between clinical trials.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-10. PROTEIN PROFILING OF CEREBROSPINAL FLUID FROM PEDIATRIC DIPG AND BSG

Amanda Saratsis 1, Sridevi Yadavilli 2, Javad Nazarian 2

Abstract

Diffuse intrinsic pontine glioma (DIPG) is a leading cause of brain tumor death in children due to lack of effective therapy. Proteomic analysis of cerebrospinal fluid (CSF) from children with DIPG is ideal for detecting characterizing tumor-associated proteins. We explored the CSF protein profile of DIPG patients to identify and validate tumor-specific proteins in CSF, tissue, serum and urine, and correlated to clinical course and radiographic imaging. CSF samples were submitted for MS/MS proteomic analysis using LTQ-Orbitrap-XL. Isolated peptides identified using the Sequest algorithm in the Bioworks browser against the Uniprot database were submitted to quantitative and subgroup analysis using ProteoIQ and Partek Genomics Suite. Proteins of interest were validated with western blot and immunohistochemical staining of CSF, brain tissue, serum and urine. Clinical course and radiologic imaging was reviewed. Molecular pathways were explored using Ingenuity Pathway Analysis. Spectral count analysis revealed selective upregulation of Cyclophillin A (CypA) in BSG patients (n = 20) compared to supratentorial glioma (n = 17) and controls (n = 22). Findings were validated using western blot analysis of CSF, serum and urine. Immunohistochemical staining revealed selective upregulation of tumor secreted CypA, but not cytosolic. CypA expression correlated with tumor grade and clinical course, and connects to potential tumorigenic pathways. We use the first comprehensive protein profile of CSF specimens from patients with BSG to demonstrate selective expression of tumor proteins potentially involved in brainstem gliomagenesis. Facile detection of tumor indicator proteins in serum and urine could have potential clinical correlation, with implications for prognosis and treatment of DIPG.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-11. HEDGEHOG PATHWAY INHIBITION ERADICATES THE TUMOR-INITIATING CELL POPULATION IN DIPG, BUT DOES NOT PROLONG SURVIVAL

Michelle Monje 1, Morgan Freret 2, Siddhartha Mitra 1, Shahid Mallick 3, Jynho Kim 1, Philip Beachy 1

Abstract

Diffuse Intrinsic Pontine Glioma (DIPG) is a devastating childhood cancer. We have recently developed the first available cell culture and orthotopic xenograft models of DIPG. These invaluable resources, together with our parallel studies of normal brainstem development, have led to the discovery of a DIPG tumor-initiating cell type. The tumor-initiating cell, or “cancer stem cell” (CSC), of DIPG is a small subpopulation that is responsive to the powerful signaling pathway Hedgehog. We have now targeted the DIPG tumor-initiating cell population in vivo using pharmacological Hedgehog pathway inhibition. While Hedgehog pathway inhibition has dramatic effects on tumor cell self-renewal in vitro, and slows tumor growth for about a month in vivo, it does not change the ultimate size of the tumors that diffusely infiltrate the brainstem in our DIPG orthotopic xenograft model nor does it significantly alter survival. Hedgehog inhibitor therapy does, however, eliminate the tumor-initiating cell population. Recovered xenografts from Hedgehog inhibitor-treated mice demonstrate eradication of cells expressing brain tumor initiating cell markers CD133 and CD15, loss of neurosphere-forming capacity in vitro and loss of tumorigenicity in serial orthotopic xenotransplantation experiments. The classic CSC hypothesis posits that the CSC is solely responsible for tumor propagation. Without the CSC, all daughter cell types would be expected to burn out and the tumor to become indolent. Following Hedgehog inhibitor therapy, the “CSC” is gone, but the tumor is still lethal for the mouse. It may be that DIPG tumor cell behavior does not conform precisely to the CSC hypothesis, or it may be that in DIPG the time required for transit amplifying cell “burn out” is not tolerated by the functionally critical, relatively small and confined brainstem. Importantly, these findings point to a paradigm shift from the classic hierarchical cancer stem cell hypothesis, and highlight the need for a novel, combinatorial strategy.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-12. PEDIATRIC BRAINSTEM GLIOMA: EXPERIENCE OF ONE BRAZILIAN INSTITUTION

Liana Nobre 1, Flavia Vasconcelos 1, Fernanda Lima 1, Debora Mattos 1, Nathalia Kuiven 1, Gabriela Lima 1, Joseane Silveira 1, Marina Sevilha 1, Marco Antonio Lima 1, Sima Ferman 1

Abstract

BACKGROUND: Brainstem tumors account for 10-20% of pediatric brain tumors. Despite various treatment approaches the prognosis is dismal. However, the unfavorable prognosis depends on extension and location. OBJECTIVE: To describe patient characteristics and treatment response of a cohort of children treated at the Instituto Nacional de Câncer with brainstem tumors. METHODS: Retrospective analysis of 48 patients admitted between January 1st, 2005 and December 31st, 2010. All patients were initially accessed with history, physical examination, brain MRI and neurosurgery consultation. The patients were classified and treated according to tumor extension and location with surgery and/or radiotherapy. RESULTS: There were 26 female and 22 male, the median age at diagnosis was 6y 6m (range: 7m-17y3m), median time from the beginning of symptoms to diagnosis was 36 days (range: 02-730 days). Classification by extension was: diffuse gliomas n = 40 (83.3%), focal n = 5 (10.4% ) exofitic n = 3 (6,3%); by location: midbrain n = 5 (10.4%), pons n = 39 (81.3%) medulla n = 4 (8.3%). Eleven patients underwent neurosurgical procedures: biopsy n = 7 and partial resections n = 4. Histopathological results were available in 10 patients, nine low grade gliomas and 1 PNET. One biopsy was inconclusive. Conventional radiotherapy with 54 Gy was given to 37/48 pts (77.1%). Four patients were treated with chemotherapy, 6 observation only and 2 patients surgery then observation. The 1-year and 2-year OS for the whole group was 49.8% and 28.6%, respectively. The 1-year and 2- year OS by location was : midbrain (100% and 100%), Pons (41% and 12.93%) and medulla (75% and 75%). The 2-year OS by extension: focal, exofitic and diffuse was 100 %, 67% and 15%, respectively. CONCLUSION: Good results were obtained in focal and exofitic tumors with standard treatment. Diffuse pontine gliomas had the worst outcome. New approaches are needed to improve the results of this group of patients.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-13. MULTICENTRIC PHASE I STUDY OF CILENGITIDE (CGT) IN COMBINATION WITH RADIATION THERAPY IN CHILDREN WITH NEWLY DIAGNOSED DIFFUSE INTRINSIC PONTINE GLIOMA: PRELIMINARY RESULTS

Pierre Leblond 1, Amélie Lansiaux 1, Xavier Rialland 2, Jean-Claude Gentet 3, Birgit Geoerger 4, Didier Frappaz 5, Isabelle Aerts 6, Valérie Bernier-Chastagner 7

Abstract

BACKGROUND: Despite numerous clinical trials, the prognosis of children with diffuse intrinsic pontine glioma (DIPG) is poor. Cilengitide (EMD121974, Merck KGaA, Germany) is a selective antagonist of αvβ3 and αvβ5 integrins known to be involved in tumor growth, cell adhesion, angiogenesis and metastasis development. Integrins are highly expressed in adult GBM in which CGT is currently evaluated. There is a lack of data concerning CGT in combination with radiation therapy (RT) in children. Therefore, we conducted a phase I study in children with newly diagnosed DIPG to determine the maximum tolerated dose (MTD), dose limiting toxicities (DLT) and pharmacokinetics (PK) of CGT, and to estimate the expression level of αvβ3 and αvβ5 integrins. METHODS: Cilengitide was intravenously administered twice weekly in a dose escalation manner following a 3 + 3 classical design, in combination with RT (Total dose 54 Gy, 1.8 Gy/fraction) during 6 weeks, then as a single agent up to tumor progression. Five dose were tested (240, 480, 720, 1200, 1800 mg/m2). PK analyses were performed (not currently available). Dose-limiting toxicities (DLTs) were evaluated during the first 6 weeks. RESULTS: These preliminary results concern the four first dose levels. Fourteen patients with 1 WHO grade II, 8 grade III, 3 grade IV, 2 not determined, were enrolled. Twelve patients were evaluable for toxicity with 3 patients at each dose level, receiving a mean of 4.2 cycles (1-8) of treatment. Median age was 6.2 years (3-15). No DLT occurred in the four first dose levels. Thirteen grade 2-5 adverse events occurred in 8 patients (hydrocephalus, neurological deterioration), but considered as tumor progression and not related to treatment. CONCLUSIONS: CGT seems to be well tolerated when used in combination with radiation therapy in children with DIPG. Final results containing data of the fifth dose level will be presented during the meeting.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-14. DISTANT METASTASIS POST VEGF-TARGETED THERAPY IN TWO PATIENTS WITH DIFFUSE INTRINSIC PONTINE GLIOMA (DIPG)

Rachana Shah 1, Wafik Zaky 1, John Grimm 1, Stefan Bluml 1, Kenneth Wong 1, Girish Dhall 1

Abstract

DIPGs account for 80% of gliomas located in the brainstem and have a universally grim prognosis. The standard of care is radiation therapy (XRT); however, effects are usually transient with a reported median survival of less than one year. Glioblastoma Multiforme (GBM) is the most common histology found in DIPGs at autopsy. Vascular endothelial growth factor (VEGF) receptor pathway activation has been demonstrated in GBM. Bevacizumab (BEV), a VEGF monoclonal antibody, is being increasingly utilized in the frontline management of GBM and DIPG. Recently, preclinical studies have demonstrated an adaptive-evasive response involving augmented tumor cell invasion and increased dissemination following anti-VEGF therapy. Here, we report two DIPG patients who developed distant metastasis after treatment with a BEV containing regimen. Patient #1 was nine years of age at diagnosis (anaplastic astrocytoma on biopsy). He received focal XRT with carboplatin and oral etoposide at diagnosis followed by irinotecan, temozolomide (TMZ) and BEV. Two months after starting BEV, he developed a metastatic lesion in the frontal horn of right lateral ventricle. He expired 15months from diagnosis with a histologic confirmation of GBM at both sites on autopsy. Patient #2 was diagnosed with DIPG at nine years of age. He received focal XRT with TMZ and vorinostat at diagnosis followed by dasatinib, lenalidomide and TMZ for two months. Following local disease progression, treatment was changed to irinotecan, TMZ and BEV. He developed distant metastasis in the frontal horn of left lateral ventricle four months after initiating BEV. He is still alive twelve months from diagnosis. To the best of our knowledge, development of distant metastasis following anti-VEGF therapy has never been reported in the DIPG literature. Further clinical trials are warranted to confirm tumor behavior and outcomes post VEGF-targeted therapy, until which BEV should be used with caution in patients with DIPG.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-15. HUMAN DIPG CELLS INJECTED IN THE MURINE PONS GIVE RISE TO MURINE TUMORS WITH PONTINE PRECURSOR-LIKE CELL MARKERS

Viola Caretti 1, Pepijn Schellen 1, Tonny Lagerweij 1, Marianna Bugiani 2, Anneke Navis 3, Pieter Wesseling 2, W Peter Vandertop 4, David P Noske 5, Gertjan Kaspers 6, Thomas Wurdinger 5

Abstract

BACKGROUND: Diffuse Intrinsic Pontine Glioma (DIPG) is a dismal pediatric malignancy, characterized by its specific anatomical location in the pons. In this study, we aimed to develop DIPG mouse models for translational research employing primary human DIPG cells. METHODS: Straight after autopsy of three pediatric DIPG patients, single cell suspensions were directly injected subcutaneously or into the pons of SCID mice (n = 3). RESULTS: In all attempts, the DIPG cells injected in the pons gave rise to tumors (VUMC-DIPG-3/4/5) with similar histopathological features as the original human DIPG tissue. In contrast, the DIPG cells injected subcutaneously never gave rise to tumors. Surprisingly, the tumors developed after injection of the human DIPG cells into the murine pons were of mouse and not of human origin. The murine VUMC-DIPG-3/4/5 tumors were negative for human Vimentin and for FISH of centromere 1 (mouse chromosomes do not have centromeres), whereas Ki67 antibodies directed specifically against the mouse antigen did react, in contrast to human E98-FM DIPG tumors. In addition, via subsequent transplantations two DIPG mouse models were established (VUMC-DIPG-3/5). The VUMC-DIPG-3/5 cells were karyotyped showing only mouse metaphases. Recently Monje et al. described a human Nestin +/Olig2 +/GFAP − and a corresponding mouse Olig2 +/SOX2 + pontine precursor like cell (PPC) population. The mouse VUMC-DIPG-3/5 tumors did reveal a Nestin +/GFAP − and Olig2 +/SOX2 + phenotype, indicative of a PPC-like DIPG component. Finally, aCGH analysis revealed a similar aberration pattern in the murine VUMC-DIPG-3/5 models, indicating a possible common pathogenesis. CONCLUSION: We demonstrate that injection of human DIPG cells into the murine pons resulted in development of murine tumors resembling the human tumour phenotype. Since this phenomenon occurred in three out of three cases, the onset of mouse DIPG tumors might be due to transmissible tumorigenic factors present in human DIPG tissue, which remain to be identified.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-16. LONGER DURATION OF SYMPTOMS AT DIAGNOSIS IS AN INDEPENDENT PROGNOSTIC FACTOR IN PEDIATRIC DIFFUSE PONTINE INTRINSIC GLIOMAS

Hannah Lee 2, David Ziegler 1

Abstract

BACKGROUND: Diffuse intrinsic pontine gliomas (DIPGs) are an almost invariably fatal childhood brain tumor. The diagnosis is based on clinical and radiological appearance and biopsy is rarely performed. While median survival is less than 10 months, a small number of patients will survive for more than 1 year. We hypothesized that some patients may have lower grade tumors that predispose to improved survival, and that there may be clinical factors that can be used to assess prognosis. The aim of this study was to determine prognostic indicators in children with DIPG. PATIENTS AND METHODS: We retrospectively reviewed the charts of 28 pediatric patients admitted to a single institution with a diagnosis of DIPG between 1990 and 2008. We recorded demographic and clinical variables, as well as radiological findings and survival. We used Cox proportional hazard models to determine prognostic factors. RESULTS: 28 patients were identified with DIPG: 15 male and 13 female. The median age at diagnosis was 5.6 years (range 3.2 - 14.2). The overall survival for all patients at 6 months, 1 year and 2 years was 68%, 36% and 11%, respectively, with a median survival of 7.9 months (range 0.3 - 109.2). Univariate analysis identified five favourable prognostic factors: younger age at diagnosis (< 5 years); duration of symptoms > 3 months; absence of necrosis on MRI; low-grade histology, and treatment with radiotherapy. In multivariate analysis, only two favourable prognostic factors were identified: duration of symptoms at diagnosis of greater than 3 months, and treatment with radiotherapy. CONCLUSIONS: The data shows that longer duration of symptoms prior to diagnosis (> 3 months) is associated with improved survival in children with DIPG. Further prospective studies will further assess the importance of these factors and ultimately help stratify patients in future clinical studies.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-17. HIGH THROUGHPUT IN VITRO DRUG SCREEN TO IDENTIFY NOVEL THERAPEUTIC TARGETS FOR DIFFUSE INTRINSIC PONTINE GLIOMA

Kristin Schroeder 1, Elaine Huang 2, Noah Berlow 3, Ranadip Patel 3, Oren Becher 1

Abstract

BACKGROUND: Diffuse intrinsic pontine glioma (DIPG) constitutes 15% of pediatric brain tumors and is the leading cause of death in children with brain tumors. Due to small patient numbers and the large number of potential therapeutic agents, there is a need for preclinical models that can help prioritize the translation of promising compounds into clinical trials for children with DIPG. Here neurospheres derived from a genetic mouse model of DIPG were used to carry out a high throughput drug screen to identify potentially novel therapeutic targets. This study was conducted as part of the Children's Oncology Group CNS-DVL committee preclinical consortium. METHODS: Two PDGF-driven DIPG mouse models were used where PDGFA overexpression and p53 loss are induced in either nestin or GFAP expressing cells of the neonatal brainstem. Neurospheres derived from the latter model were used in an unbiased high throughput drug screen with 60 known drug compounds. IC50 values were determined using MTS or Cell Glo viability assays. Promising compounds and their targets were validated with repeat assays and western blot analysis, respectively. RESULTS: Six drugs were found to have significant cytotoxic effects in two separate assays at clinically achievable concentrations: BMS 754807 (IGF-1R), carfilzomib (proteasome), obatoclax (Bcl2), SNS-032 (CDK2, 7, 9), crizotinib (ALK/c-MET) and Bix 01294 (histone-lysine methyltransferase). On repeat assays with both models, SNS-032 and BMS 754807 were found to have greater than 80% cytocidal activity, with IC50 less than 1 µM concentration. Western blot target validation was performed for SNS-032. A probabilistic Boolean model identified CDK 2 as a recurring target of interest. CONCLUSION: Using a high throughput in vitro drug screen on neurospheres derived from a genetic mouse model of DIPG can help prioritize therapy evaluation. Current results identify CDK2, 7, 9 as potential important targets for future in vivo testing and clinical translation.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-18. A HUMAN NEURAL STEM CELL BASED MODEL FOR DIFFUSE INTRINSIC PONTINE GLIOMA

Isabella Taylor 1, Xing-gang Mao 1, Marianne Hutt 1, Melanie Weingart 1, Ulf Kahlert 2, Jarek Maciacyk 3, Guido Nikkhah 2, Charles Eberhart 1, Eric Raabe 1

Abstract

Diffuse intrinsic pontine glioma (DIPG) is a highly fatal pediatric brain tumor of the brain stem. The few studies of the genetics of DIPG have shown that there are common deletions or mutations in the tumor suppressors TP53 and PTEN. AKT amplifications are also observed in a subset of DIPG, suggesting that this pathway is highly important to tumorigenesis. DIPG samples are also reported to contain c-MYC amplification, and MYC is a known oncogene in glioblastoma. hTERT is also known to be amplified in DIPG, and is commonly expressed in many tumors. We hypothesized that a combination of these oncogenes would lead to transformation of human hindbrain-derived neural stem and progenitor cells. To test this hypothesis, we used lentivirus to transduce these cells with sequential oncogenic alterations. Oncogenic elements introduced included constitutively active AKT, dominant negative R248W p53, c-MYC, and hTERT. These oncogenes were added to neural stem and progenitor cells in different combinations. After antibiotic and GFP selection, we verified by western blotting that the resulting cells contained the appropriate oncogenes. In vitro studies showed that c-MYC, R248Wp53/c-MYC/hTERT and R248Wp53/c-MYC/hTERT/AKT cells proliferated at an increased rate compared to control hindbrain stem and progenitor cells. We performed stereotactic hindbrain injections in immunocompromised mice. Tumors formed in three months in MYC/R248Wp53/hTERT transduced cells. Cells transduced with MYC/R248Wp53/hTERT/AKT formed tumors in one to two months. Immunohistochemical analysis showed that these tumors were GFAP negative, synaptophysin positive and had PNET histology. Hindbrain cells transduced with c-MYC only and p53/hTERT did not form tumors. We conclude that packages of oncogenes can transform human hindbrain-derived neural stem and progenitor cells. This model system will be of value to investigate the significance of potential driver mutations in DIPG.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-19. TREATMENT OF BRAINSTEM GLIOMAS WITH A CDK 4/6 INHIBITOR USING GENETICALLY ENGINEERED MOUSE MODELS

Kelly Barton 1, Katherine Misuraca 1

Abstract

High-grade brainstem glioma (BSG) accounts for 15% of pediatric brain tumors and is the leading cause of death in children with brain tumors. Despite numerous clinical trials evaluating chemotherapy and novel targeted agents, the natural history of this disease has not been significantly affected and 90% of children with high-grade BSG die within 2 years of diagnosis. Recent genomic studies demonstrated that 30% of pediatric high-grade BSGs harbor amplification of PDGFRa, 40% harbor p53 mutations, and 30% have focal gains in cell-cycle control proteins: CDK4, CDK6, Cyclin D1, D2, or D3. We recently developed a genetically engineered mouse model of brainstem glioma (BSG) that is driven by over-expression of PDGF-B in Nestin expressing cells residing in the brainstem of Ink4a-ARF deficient or p53 deficient neonatal mice. Here we evaluated the efficacy of a CDK4-6 inhibitor (PD-0332991) in both the Ink4a-ARF deficient model and the p53 deficient model. In vitro, PD-0332991 induced cell-cycle arrest with IC50s around 50-500nM in cell-lines derived from the Ink4a-ARF deficient BSG model but it did not induce cell-cycle arrest at doses up to 2uM in a cell-line derived from the p53 deficient BSG model. In vivo, two daily doses of PD-0332991 induce significant cell-cycle arrest in the Ink4a-ARF BSG model (p = 0.0047). In addition, as recent genomic evidence of human pediatric gliomas suggest Ink4a-ARF deletions are more common in pediatric cortical gliomas as opposed to BSGs, we also tested PD-0332991 in a PDGF-B driven Ink4a-ARF deficient cortical glioma model in vivo and observed that a single dose can significantly inhibit proliferation (p = 0.0286). We are currently evaluating the short-term efficacy of PD-0332991 in vivo in the p53 deficient BSG model and evaluating whether treatment with PD-0332991 will provide a significant survival benefit in the Ink4a-ARF deficient BSG model.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-20. REGIONAL DIFFERENCES IN GLIOMAGENESIS: BRAINSTEM VERSUS CORTICAL GLIOMA AND THE ROLE OF PAX3

Katherine Misuraca 1, Oren Becher 1

Abstract

Brainstem Glioma (BSG), also known as diffuse intrinsic pontine glioma (DIPG), is a rare and incurable pediatric brain tumor with less than a 1-year median survival rate due to a lack of therapeutic options. Discrepancies in the clinical behavior of DIPG when compared with cortical gliomas (CG), along with biological data suggesting regional differences in neural stem and progenitor cells, led us to hypothesize that gliomas arising in different parts of the brain are biologically distinct from one another and a better understanding of these differences could lead to improved therapies for DIPG. Using the RCAS/tv-a avian retroviral system, we are able to model PDGFB-driven BSG and CG by targeting Nestin-expressing cells of the brainstem and cortex, respectively, of Ink4a/ARF- or p53-deficient mice. Expression profiling analysis of PDGFB-induced mouse BSGs versus CGs uncovered significant upregulation of the transcription factor pax3 in BSG. Analysis of PDGFB;Ink4aARF − /-;Olig2GFP tumors revealed that pax3 uniquely characterizes the Olig2-tumor cell compartment in BSG. Overexpression of Pax3 in Nestin-expressing cells of wildtype brainstem enhances PDGFB-induced tumorigenesis in vivo, decreasing overall survival (p = 0.01), and increasing tumor incidence from 36.4% in mice receiving PDGFB alone (n = 11) to 81% in mice receiving PDGFB and Pax3 (n = 16). The ectopic expression of Pax3 also induced higher-grade malignancies, with 3/13 tumors showing characteristics of grade III gliomas (compared to 0/4 of the PDGFB-induced tumors). These data suggest that Pax3 plays a functional role in brainstem gliomagenesis. The differential expression pattern of Pax3 in the mouse is paralleled in human pediatric glioma, with pax3 mRNA expressed at significantly higher levels in BSGs than in CGs (p = 0.006). High Pax3 expression characterizes roughly one-third of human DIPG; we hope that further elucidation of the mechanism of Pax3 function will lead to new therapeutic options for these patients.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-21. IN VITRO ASSESSMENT OF TYROSINE KINASE INHIBITOR TREATMENT FOR DIFFUSE INTRINSIC PONTINE GLIOMA

Zhiping Zhou 1, Lauren Rotman 2, Sharon Ho 1, Mark Souweidane 1

Abstract

INTRODUCTION: Diffuse intrinsic pontine glioma (DIPG) in childhood is a uniformly lethal condition with a median survival of only 8-10 months. There have been no chemotherapeutic or biological agents that have proven effective in improving the survival of children with DIPG. Recent molecular characterization has revealed that platelet-derived growth factor receptor (PDGFR), a receptor tyrosine kinase, is overexpressed and activated in the majority of DIPG cases. Systemic administration of a PDGFR inhibitor (imatinib) has shown minimal efficacy in limited clinical cases. We investigated using tyrosine kinase inhibitors (TKIs) as a therapy for DIPG and potential ways to improve their efficacy. METHODS: Brainstem glioma (BSG) cells were grown from an RCAS-PDGF-B-driven mouse DIPG model. Experimental therapy was performed using dasatinib, a TKI with strong inhibitory effects on PDGFR. Efficacy was assessed by cell growth and survival. Immunocytochemistry and Western blot was performed to assess expression and phosphorylation status of certain signal transduction molecules. RESULTS: Cell growth was decreased to 10.1% with dasatinib (100nM, 72 hours) treatment compared with control. Akt phosphorylation was incompletely inhibited by dasatinib treatment. Adding perifosine (30 µM), an Akt inhibitor, to the regimen further decreased cell growth to 5.1% while perifosine itself reduced cell growth to 29.7%. Further analysis showed that autophagy signaling was activated within 2 hours of dasatinib treatment and apoptosis was activated after 8 hours of treatment. CONCLUSION: TKI does not completely inhibit downstream survival signals, indicating these downstream signals are valid targets for combination therapies. The activation of both autophagy and apoptosis suggests TKIs have complex effects on PDGF-driven BSG cells that need to be elucidated to improve therapeutic efficacy.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-22. BLOCKADE OF THE NOTCH PATHWAY IN DIFFUSE INTRINSIC PONTINE GLIOMA

Marianne Hutt 1, Kah Jing Lim 1, Katherine Warren 3, Howard Chang 2, Charles Eberhart 1, Eric Raabe 1

Abstract

Diffuse intrinsic pontine glioma (DIPG) is a universally fatal pediatric brain tumor. As part of the Mid-Atlantic DIPG Consortium, three institutions (Johns Hopkins, Children's National Medical Center, and the National Institutes of Health - Pediatric Oncology Branch) are collaborating to collect and disseminate rapid-autopsy material from patients with DIPG. We have established a human DIPG xenograft and neurosphere cell line from a rapid autopsy specimen. In vitro, these cells are GFAP positive, and express markers of human neural stem cells and glioblastoma, including human specific SOX2 and NESTIN. They also show marked invasive properties in vitro as measured by transwell assay. In vivo JHH DIPG1 forms tumor xenografts in approximately six months in immunocompromised animals. These tumor cells diffusely infiltrate distant brain structures, recapitulating the invasive phenotype of DIPG. Due to their high level expression of neural stem cell markers, and the well-known roles of the Notch pathway in stem cell biology and aggressive brain tumors, we examined Notch signaling in DIPG. Notch 1 and 3 and Notch downstream effectors, specifically Hes1, Hes5 and Hey1, are highly expressed in JHH DIPG1 and a second DIPG cell line, SU DIPG1. Notch pathway activation is equal to or exceeding that observed in glioblastoma neurosphere cell lines. Notch targeting using gamma-secretase inhibitors leads to decreased cell proliferation in a dose-dependent fashion. These lines represent two of only a handful of pediatric DIPG cell lines extant, and the high level of Notch pathway activation in these cells suggests that blockade of Notch pathway using gamma secretase inhibitors and shRNAs against canonical Notch effector CBF may be efficacious in DIPG. Our laboratory is actively investigating the role of the Notch pathway in DIPG in pre-clinical studies in mice, with the goal of moving Notch inhibition into clinical trials for these currently highly lethal tumors.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-23. BRAINSTEM PRIMITIVE NEUROECTODERMAL TUMORS MIMICKING DIFFUSE INTRINSIC PONTINE GLIOMAS.

Donita Lightner 1, Sophia Haque 1, Mark Souweidane 2, Yasmin Khakoo 1, Ira Dunkel 1, Stephen Gilheeney 1, Kim Kramer 1, David Lyden 1, Suzanne Wolden 1, Jeff Greenfield 2, Kevin De Braganca 1

Abstract

INTRODUCTION: Brainstem primitive neuroectodermal tumors (PNETs) are extremely rare. These tumors resemble diffuse intrinsic pontine glioma (DIPG) on MRI. We report two patients with brainstem PNETs and critical neuroimaging findings. CASE REPORT: Patient 1 presented at age five years with diplopia and ataxia. MRI revealed a noncontrast-enhancing pontine lesion with restricted diffusion. A subtotal resection demonstrated PNET. She was treated as per ACNS0332. She is almost two years out from diagnosis and has no evidence of disease. Patient 2 presented with diplopia and hemiparesis at age 23 months. MRI revealed a noncontrast-enhancing lesion with restricted diffusion of the pons. She was diagnosed with DIPG and treated with standard radiation. Four months after treatment, biopsy was performed and revealed necrotic tissue of indeterminate significance. She was referred to our institution for further management and observation was elected. One month after biopsy, she presented with worsening symptoms. Repeat MRI showed interval enlargement of a noncontrast-enhancing lesion with restricted diffusion. Subtotal resection revealed PNET. She will be treated as per ACNS0334. DISCUSSION: The pertinence of diffusion weighted imaging (DWI) sequences in brainstem lesions has limited recognition. A recent analysis by the Pediatric Brain Tumor Consortium (PBTC) of MRI's in patients with brainstem gliomas noted a longer progression free survival and overall survival in patients with a decrease in tumor volume and diffusion values. Considering these findings, DIPG patients with restricted diffusion may warrant biopsy. These cases highlight the importance of DWI sequences in brainstem tumors and potentially impact management.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-24. COMBINED BEVACIZUMAB AND TEMOZOLOMIDE ON PEDIATRIC BRAIN STEM PONTINE GLIOMAS

Hsu Ting-Rong 1, Liang Muh-Li 1, Chang Kai-Ping 1, Wong Tai-Tong 1, Chen Hsin-Hung 1

Abstract

BACKGROUND: The treatment for pediatric brain stem gliomas is very difficult and the prognosis is very poor. There was little data about combined bevacizumab and temozolomide on pediatric brain stem pontine gliomas. METHOD: We enrolled 3 children with brain stem pontine tumor since 2010. Combined radiotherapy and temozolomide was used, and then followed by combined intravenous bevacizumab and oral temozolomide. The progression free survival, survival time and side effect were observed. RESULT: There are two girls and one boy with brain stem glioblastoma multiforme, which was diagnosed in 2010. All three are 10 years old. They received combined radiotherapy and oral temozolomide treatment. Then combined bevacizumab and temozolomide was used. All of the patients showed tumor recurrence seven, eleven and fourteen 14 months later. The two patients expired after ten and sixteen months and one survived about 19 months till now. One girl who expired suffered from acute stroke when received 22 courses of bevacizumab. CONCLUSION: We used combined bevacizumab and temozlomide in treatment for 3 pediatric brain stem glioblastoma multiforme. The progression free survival and survival time showed a little prolonged. One had adverse effect of bevacizumab and acute stroke attacked. Further large study was need for evaluation of combined bevacizumab and temozolomide in treatment of pediatric brain stem pontine gliomas.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-25. DIFFUSE INTRINSIC PONTINE GLIOMAS (DIPG) IN CHILDREN: RESULTS FROM A SINGLE CENTER

Rejin Kebudi 1, F Betul Cakir 2, Fulya Y Agaoglu 3, Ömer Görgün 1, Yavuz Dizdar 3, Inci Ayan 3, Emin Darendeliler 3

Abstract

BACKGROUND: The prognosis of children with DIPG is dismal. Despite various studies undertaken to improve outcome, radiotherapy (RT) remains the standard treatment, which is mostly palliative. This study aims to evaluate characteristics and treatment outcome of children with DIPG in a single center. METHODS: We retrospectively reviewed the characteristics and treatment outcome of children with DIPG treated at Istanbul University, Oncology Institute between1999-2011. We also evaluated the group that prospectively recieved RT with concurrent and adjuvant temozolamide after 2004. RESULTS: 47 children (24 female, 23 male), median age 7 years (6 months-16 years) were analyzed. The median duration of symptoms was 30 days (2-630 days). The frequent clinical findings were ataxia, strabismus, motor weakness. All patients received RT, 54-60 Gy to the tumor site. 12 recieved only RT. 35 had concomitant and/or adjuvant chemotherapy with RT. Since 2004, 20 patients recieved the institutional protocol- temozolomide (TMZ) (75 mg/m2/day) for 6 weeks concurrent with RT, followed by TMZ (200 mg/m2/day) for 5 days/every 28 days for 12 cycles or until progression. There was no major side effect due to TMZ. The median overall survival after diagnosis was 13 months (3-132 mo.) for the whole group. The median overall survival in 20 patients that received RT and TMZ [17 months (3-132 months)], was significantly superior than that in 12 patients that recieved only RT [12 months (3-20 months)] ((p = 0.03). Nimotuzumab was given to 4 patients that progressed after RT and TMZ. There was no major side effect due to nimotuzumab. One was stable for 1 year with significant clinical improvement, the others were stable for 5, 2 and 2 months after nimotuzumab. CONCLUSIONS: In our series, the median survival was significantly superior in patients who received RT with concurrent and adjuvant temozolamide in comparison to patients that recieved RT alone.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-26. CHEMOTHERAPY ADDED TO RADIOTHERAPY FAILED TO IMPROVE DISMAL OUTCOME OF CHILDREN WITH DIPG: SINGLE INSTITUTION EXPERIENCE

Michal Zapotocky 1, Marketa Churackova 1, Bela Malinova 2, Roman Kodet 3, Martin Kyncl 4, Michal Tichy 5, Jan Stary 1, David Sumerauer 1

Abstract

Despite the significant progress in anticancer drug development diffuse intrinsic pontine glioma (DIPG) represents the diagnosis with dismal prognosis. To evaluate outcome of children with DIPG in our institution we analyzed 21 patients who were diagnosed and treated within years 2000-2011. Patients cohort consisted of 14 boys and 7 girls with median of age 5.8 years (range 0.6 to 18.8 years), with only one patient of age under 3 years. Parents or guardians of 4 patients refused any treatment, 1 patient received chemotherapy only (carboplatin and vincristine) because of low age (6 months). Radiotherapy 59.4Gy was administered alone (n = 3), or followed by vincristine, CCNU and prednisone (n = 7) or concomitantly with temozolomide (TMZ) followed by maintenance therapy with TMZ (n = 6). One patient received additional CCNU together with TMZ regimen. Only two patients were biopsied at the time of diagnosis with the result of anaplastic astrocytoma gr. III. Probability of 1-year OS of patients who received any therapy was 35.2% with median survival time 10.4 months (range 2.2 - 32.4 months), while all 4 patients without treatment died during the first year with median of survival 2.4 months (range 1.9-11.1 months). There were no significant differences in survival among patients treated by TMZ or VCR/CCNU/prednisone or radiotherapy only. The only one long term survivor is patient diagnosed at 6 months of age (2.7 years). Although our data are based on low numbers of patients we have shown that the poor outcome cannot be improved by any chemotherapy we used, leaving radiotherapy as a standard of palliative treatment. Novel strategies based on results from currently published whole-genome sequencing and other molecular biology studies should be considered for future directions in treatment of pediatric DIPG. Supported by OPPK CZ.2.16/3.1.00/24022 and Institutional science support of the University Hospital Motol

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-27. PHASE I TRIAL OF CONCURRENT WEEKLY PACLITAXEL AND RADIATION THERAPY FOR CHILDREN WITH NEWLY DIAGNOSED DIFFUSE INTRINSIC PONTINE GLIOMA

Jane Minturn 1, Hui-Kuo Shu 2, Michael Fisher 1, Ratnakar Patti 1, Anna Janss 2, Jeffrey Allen 3, Peter Phillips 1, Jean Belasco 1

Abstract

PURPOSE: Children diagnosed with diffuse intrinsic pontine glioma (DIPG) have a dismal prognosis with a median overall survival of 8-12 months. Radiation therapy (RT) is considered standard treatment, but is essentially palliative. Paclitaxel inhibits tumor growth by promoting mitotic arrest and the induction of apoptosis. It also acts as a cell-cycle specific radiosensitizer. METHODS: We performed a phase I trial of weekly paclitaxel with concurrent RT in patients with newly diagnosed DIPG. Subjects received intravenous paclitaxel 1-3 hours prior to RT, weekly for six weeks. Paclitaxel starting dose was 175 mg/m2/dose and was escalated using a 3 + 3 design. Radiation was delivered in standard 180cGy fractions to a dose of 54Gy. Response data were obtained six weeks after completion of RT by MRI scan. Subjects could receive an additional 8 courses of post-RT paclitaxel. RESULTS: Eleven patients were enrolled (median age 8.3 years; range, 3-17), and three dose levels explored. All patients were evaluable for dose escalation and response. Two of five patients treated at dose-level 3 (225 mg/m2/dose) experienced grade 4 toxicities. The median time from diagnosis to disease progression was 6.4 months, and the median time from diagnosis to death was 9.9 months. One subject had a partial response following RT and stable disease 106 months from diagnosis. There was subsequent disease progression, but she remains alive 10.9 years from diagnosis. A recommended phase II dose of 200mg/m2/dose during RT was established. CONCLUSION: Delivery of infusional paclitaxel to children with DIPG as a radiosensitizer was feasible and tolerated at 200 mg/m2/dose weekly during involved-field RT. The study was closed early because of slow accrual due to availability of competing protocols and little evidence of activity. The addition of paclitaxel to radiation did not appear to prolong life in subjects with DIPG beyond that with standard radiation alone.

Neuro Oncol. 2012 Jun;14(Suppl 1):i26–i32.

PG-28. ESTABLISHMENT OF AN INTERNATIONAL REPOSITORY OF DIPG GENOMICS DATA

Kathryn Taylor 1, Michael Baudis 2, Andre von Beuren 3, Maryam Fouladi 4, Chris Jones 1

Abstract

Diffuse intrinsic pontine glioma represents one of the biggest therapeutic challenges in paediatric neuro-oncology with a median survival of around 9 months despite collaborative efforts to improve treatment. The development of targeted therapies for DIPG has been hampered by the lack of knowledge of the biology of this devastating disease. Recent data suggest however that paediatric high-grade gliomas differ from their adult counterparts, and that there may be biological distinctions between childhood gliomas presenting in the brainstem compared with supratentorially. In order to provide a more robust assessment of these differences across as wide a range of DIPG specimens as possible, we have established an international repository of DIPG genomics data. Molecular data including, but not restricted to, genome-wide DNA copy number, karyotyping, expression profiling (mRNA and miRNA), methylation analysis and somatic mutations are gathered retrospectively through a systematic review of published literature in DIPG, carried out according to PRISMA guidelines, as well as prospectively in a collaborative manner. The repository is built using the arrayMap structure powered by the Progenetix platform (dipg.progenetix.org). Initial collaborative analyses will be undertaken with the following goals: (1) create a low-resolution fully integrated dataset to define the frequencies of chromosomal alterations; (2) retain the original platform-specific data to accurately map amplification/deletion breakpoints; (3) compare the data with publicly available paediatric and adult high-grade glioma datasets; (4) investigate any retrospective clinicopathological correlations of the genetic aberrations identified; (5) better define the intrinsic subgroups of the disease based upon gene expression signatures; and (6) integrate expression and genetic/epigenetic data with the copy number studies. We intend that by combining these data we will generate a comprehensive, accessible database of the molecular profiles of DIPG for the academic community, and welcome contributions from interested researchers.

Articles from Neuro-Oncology are provided here courtesy of Society for Neuro-Oncology and Oxford University Press

RESOURCES