Search

PURPOSE Survivors of childhood medulloblastoma suffer from substantial late effects. We characterized these sequelae using real-world health services data in a population-based cohort of medulloblastoma survivors. METHODS All 5-year medulloblastoma survivors diagnosed age < 18 years between 1987 and 2015 in Ontario, Canada, were identified and matched 1:5 with population controls. Index date was 5 years from latest pediatric cancer event. Linkage to provincial administrative health data allowed for comparison of cumulative incidences of several adverse outcomes. RESULTS Two hundred thirty survivors, 81.3% of whom had received craniospinal irradiation, were matched with 1,150 controls. The 10-year postindex cumulative incidence of all-cause mortality was 7.9% (95% CI, 3.9 to 11.8) in survivors versus 0.6% (95% CI, 0.1 to 1.1) in controls (hazard ratio [HR], 21.5; 95% CI, 9.8 to 54.0). The cumulative incidence of stroke was higher in survivors (4.8%; 95% CI, 2.2 to 9.0) compared with controls (0.1; 95% CI, 0.01 to 0.7; HR, 45.6; 95% CI, 12.8 to 289.8). Hearing loss requiring an amplification device was present in 24.9% (95% CI, 18.8 to 31.4) of survivors versus 0.3% (95% CI, 0.1 to 1.0) of controls (HR, 96.3; 95% CI, 39.7 to 317.3). Disability support prescription claims were submitted by 44.5% (95% CI, 37.1 to 51.6) of survivors versus 5.5% (95% CI, 4.2 to 7.1) of controls (HR, 10.0; 95% CI, 7.3 to 13.6). Female survivors were significantly less likely to deliver a liveborn child compared with controls (HR, 0.2; 95% CI, 0.1 to 0.7). CONCLUSION Survivors of medulloblastoma have significant long-term medical sequelae, increased all-cause mortality, and are frequently dependent on disability supports. Efforts to reduce the toxicity of current therapy, specifically incorporating molecularly informed risk stratification to spare low- and intermediate-risk survivors the toxicity of treatment, are urgently needed. These findings should prompt a re-evaluation of our current treatment approaches where research focused on late-effect interventions should be prioritized.

PURPOSE BRAF V600 mutations occur in many childhood cancers, including approximately 20% of low-grade gliomas (LGGs). Here, we describe a phase I/II study establishing pediatric dosing and pharmacokinetics of trametinib with or without dabrafenib, as well as efficacy and safety in a disease-specific cohort with BRAF V600–mutant LGG; other cohorts will be reported elsewhere. METHODS This is a four-part, phase I/II study (ClinicalTrials.gov identifier: NCT02124772 ) in patients age < 18 years with relapsed/refractory malignancies: trametinib monotherapy dose finding (part A) and disease-specific expansion (part B), and dabrafenib + trametinib dose finding (part C) and disease-specific expansion (part D). The primary objective assessed in all patients in parts A and C was to determine pediatric dosing on the basis of steady-state pharmacokinetics. Disease-specific efficacy and safety (across parts A-D) were secondary objectives. RESULTS Overall, 139 patients received trametinib (n = 91) or dabrafenib + trametinib (n = 48). Trametinib dose-limiting toxicities in > 1 patient (part A) included mucosal inflammation (n = 3) and hyponatremia (n = 2). There were no dose-limiting toxicities with combination therapy (part C). The recommended phase II dose of trametinib, with or without dabrafenib, was 0.032 mg/kg once daily for patients age < 6 years and 0.025 mg/kg once daily for patients age ≥ 6 years; dabrafenib dosing in the combination was as previously identified for monotherapy. In 49 patients with BRAF V600–mutant glioma (LGG, n = 47) across all four study parts, independently assessed objective response rates were 15% (95% CI, 1.9 to 45.4) for monotherapy (n = 13) and 25% (95% CI, 12.1 to 42.2) for combination (n = 36). Adverse event–related treatment discontinuations were more common with monotherapy (54% v 22%). CONCLUSION The trial design provided efficient evaluation of pediatric dosing, safety, and efficacy of single-agent and combination targeted therapy. Age-based and weight-based dosing of trametinib with or without dabrafenib achieved target concentrations with manageable safety and demonstrated clinical efficacy and tolerability in BRAF V600–mutant LGG.

PURPOSE Hearing loss (HL) is a serious secondary effect of treatment for CNS and head-and-neck tumors in children. The goal of this study was to evaluate incidence and risk factors for HL in patients with multiple ototoxic exposures. PATIENTS AND METHODS We evaluated 340 ears from 171 patients with CNS or head-and-neck tumors treated with radiation, with or without chemotherapy, who had longitudinal audiologic evaluation. International Society of Pediatric Oncology-Boston grades were assigned to 2,420 hearing assessments. Multivariable weighted ordinal logistic regression was fitted to evaluate the effect of clinicopathologic features on HL. RESULTS Mean cochlea dose (odds ratio [OR] 1.04 per Gy, P < .001), time since radiotherapy (RT; OR 1.21 per year, P < .001), cisplatin dose (OR 1.48 per 100 mg/m2, P < .001), and carboplatin dose (OR 1.41 per 1,000 mg/m2, P = .002) were associated with increasing International Society of Pediatric Oncology-Boston grade of HL. There was no synergistic effect of RT and cisplatin (interaction term, P = .53) or RT and carboplatin (interaction term, P = .85). Cumulative incidence of high-frequency HL (> 4 kHz) was 50% or greater at 5 years after RT if mean cochlea dose was > 30 Gy, while incidence of HL across all frequencies continued to increase beyond 5 years after RT. CONCLUSION Children treated with radiation and chemotherapy experience a high incidence of HL over time, with associations found between more severe HL and cisplatin or carboplatin dose as well as mean cochlea dose. Mean cochlea dose of ≤ 30 Gy is proposed as a goal to reduce the risk of HL; a lower threshold (20-25 Gy) may be considered in patients receiving platinum chemotherapy to reduce cumulative HL burden.

PURPOSE Constitutional mismatch repair deficiency syndrome (CMMRD) is a lethal cancer predisposition syndrome characterized by early-onset synchronous and metachronous multiorgan tumors. We designed a surveillance protocol for early tumor detection in these individuals. PATIENTS AND METHODS Data were collected from patients with confirmed CMMRD who were registered in the International Replication Repair Deficiency Consortium. Tumor spectrum, efficacy of the surveillance protocol, and malignant transformation of low-grade lesions were examined for the entire cohort. Survival outcomes were analyzed for patients followed prospectively from the time of surveillance implementation. RESULTS A total of 193 malignant tumors in 110 patients were identified. Median age of first cancer diagnosis was 9.2 years (range: 1.7-39.5 years). For patients undergoing surveillance, all GI and other solid tumors, and 75% of brain cancers were detected asymptomatically. By contrast, only 16% of hematologic malignancies were detected asymptomatically ( P < .001). Eighty-nine patients were followed prospectively and used for survival analysis. Five-year overall survival (OS) was 90% (95% CI, 78.6 to 100) and 50% (95% CI, 39.2 to 63.7) when cancer was detected asymptomatically and symptomatically, respectively ( P = .001). Patient outcome measured by adherence to the surveillance protocol revealed 4-year OS of 79% (95% CI, 54.8 to 90.9) for patients undergoing full surveillance, 55% (95% CI, 28.5 to 74.5) for partial surveillance, and 15% (95% CI, 5.2 to 28.8) for those not under surveillance ( P < .0001). Of the 64 low-grade tumors detected, the cumulative likelihood of transformation from low-to high-grade was 81% for GI cancers within 8 years and 100% for gliomas in 6 years. CONCLUSION Surveillance and early cancer detection are associated with improved OS for individuals with CMMRD.

PURPOSE Proton radiotherapy (PRT) may lessen the neuropsychological risk traditionally associated with cranial radiotherapy for the treatment of pediatric brain tumors by reducing the dose to normal tissue compared with that of photon radiotherapy (XRT). We examined the change in intellectual scores over time in patients with pediatric medulloblastoma treated with craniospinal PRT versus XRT. METHODS Intelligence test scores were obtained for a sample of pediatric patients treated between 2007 and 2018 on the same medulloblastoma protocols that differed only in radiotherapy modality (PRT v XRT). Growth curve analyses compared change in scores over time since diagnosis between groups. RESULTS Longitudinal intelligence data from 79 patients (37 PRT, 42 XRT) were examined. Groups were similar on most demographic/clinical variables, including sex (67.1% male), age at diagnosis (mean, 8.6 years), craniospinal irradiation dose (median, 23.4 Gy), length of follow-up (mean, 4.3 years), and parental education (mean, 14.3 years). Boost dose ( P < .001) and boost margin ( P = .001) differed between groups. Adjusting for covariates, the PRT group exhibited superior long-term outcomes in global intelligence quotient (IQ), perceptual reasoning, and working memory compared with the XRT group (all P < .05). The XRT group exhibited a significant decline in global IQ, working memory, and processing speed (all P < .05). The PRT group exhibited stable scores over time in all domains with the exception of processing speed ( P = .003). CONCLUSION To our knowledge, this is the first study to compare intellectual trajectories between pediatric patients treated for medulloblastoma with PRT versus those treated with XRT on comparable, contemporary protocols. PRT was associated with more favorable intellectual outcomes in most domains compared with XRT, although processing speed emerged as a vulnerable domain for both groups. This study provides the strongest evidence to date of an intellectual sparing advantage with PRT in the treatment of pediatric medulloblastoma.

Purpose Constitutional mismatch repair deficiency (CMMRD) is a highly penetrant cancer predisposition syndrome caused by biallelic mutations in mismatch repair (MMR) genes. As several cancer syndromes are clinically similar, accurate diagnosis is critical to cancer screening and treatment. As genetic diagnosis is confounded by 15 or more pseudogenes and variants of uncertain significance, a robust diagnostic assay is urgently needed. We sought to determine whether an assay that directly measures MMR activity could accurately diagnose CMMRD. Patients and Methods In vitro MMR activity was quantified using a 3′-nicked G-T mismatched DNA substrate, which requires MSH2-MSH6 and MLH1-PMS2 for repair. We quantified MMR activity from 20 Epstein-Barr virus–transformed lymphoblastoid cell lines from patients with confirmed CMMRD. We also tested 20 lymphoblastoid cell lines from patients who were suspected for CMMRD. We also characterized MMR activity from patients with neurofibromatosis type 1, Li-Fraumeni syndrome, polymerase proofreading-associated cancer syndrome, and Lynch syndrome. Results All CMMRD cell lines had low MMR activity (n = 20; mean, 4.14 ± 1.56%) relative to controls (n = 6; mean, 44.00 ± 8.65%; P < .001). Repair was restored by complementation with the missing protein, which confirmed MMR deficiency. All cases of patients with suspected CMMRD were accurately diagnosed. Individuals with Lynch syndrome (n = 28), neurofibromatosis type 1 (n = 5), Li-Fraumeni syndrome (n = 5), and polymerase proofreading-associated cancer syndrome (n = 3) had MMR activity that was comparable to controls. To accelerate testing, we measured MMR activity directly from fresh lymphocytes, which yielded results in 8 days. Conclusion On the basis of the current data set, the in vitro G-T repair assay was able to diagnose CMMRD with 100% specificity and sensitivity. Rapid diagnosis before surgery in non-neoplastic tissues could speed proper therapeutic management.

LBA2002 Background: LGG is the most common pediatric brain cancer, and BRAF V600 mutation has been detected in ≈17% of cases. Most patients (pts) with pLGG have disease progression and require post-surgical therapy. The standard of care is chemotherapy, such as carboplatin + vincristine (C+V), which may be less effective in BRAF V600–mutant disease; thus, alternative treatment options are needed. Dab + tram showed encouraging efficacy in a Phase I/II study (NCT02124772) in pts with previously treated BRAF V600–mutant pLGG. We describe the results of a randomized Phase II study (NCT02684058) of first-line dab + tram vs C+V in BRAF V600–mutant pLGG. Methods: Pts aged 1 to

2009 Background: HGGs comprise ≈10% of pediatric central nervous system tumors and are a leading cause of childhood cancer-related death. Overall response rates (ORRs) with current standards of care are low, particularly in the second line, and 2-y overall survival (OS) rates are ≤35%. BRAF V600 mutation is infrequent (≈5% of pHGGs) but associated with improved survival from time of initial diagnosis. In previous trials, dab monotherapy and dab + tram yielded encouraging outcomes in BRAF V600–mutant HGG in pediatric and adult patients (pts), respectively. We describe the results of a Phase II study (NCT02684058) of dab + tram in r/r BRAF V600–mutant pHGG. Methods: Pts aged 1 to

10053 Background: Most medulloblastoma survivors suffer from late treatment-related sequelae. There are no population-based studies examining such late effects in a dedicated cohort of medulloblastoma survivors. Methods: Using a provincial pediatric cancer registry, all 5+ year medulloblastoma survivors diagnosed between 1987-2015 at

10054 Background: Ependymoma is the third most common pediatric central nervous system tumour. Treatment approaches are intensive and may include surgery, radiation, and chemotherapy. There are no longitudinal population-based cohort studies evaluating the long-term medical and functional outcomes of survivors of childhood ependymoma. Methods: Using a provincial pediatric cancer registry, all 5+ year ependymoma survivors diagnosed between 1987-2015 in Ontario, Canada were identified and matched to cancer-free population controls based on age, sex, and geographical location. Cases were followed from the index date (5 years from latest of diagnosis, or relapse/subsequent malignancy prior to age 18 years) until December 31, 2020 or censorship (death, or relapse/ new cancer after age 18 years). Clinical data were linked to administrative health databases to estimate the cumulative incidences and cause-specific hazard ratios (HR) of mortality, hospitalizations, strokes, hearing loss requiring a hearing aid, receipt of homecare services, and subsequent malignant neoplasms (SMNs) between cohorts, accounting for matching and competing risks. Results: Of 166 ependymoma diagnoses in the study period, 70 (42.2%) were excluded, most commonly due to early death prior to the index date. Ninety-six cases were matched to 480 controls (Table). The 10-year survival probability after the index date was 92.8% in cases and 99.6% in controls (HR 9.3, 95% CI 2.3-45.2, p=0.002). Compared to controls, cases were at higher risk of hospitalization (HR 3.2, 95% CI 2.2-4.6, p

Purpose Bevacizumab (BEV) is approved in more than 60 countries for use in adults with recurrent glioblastoma. We evaluated the addition of BEV to radiotherapy plus temozolomide (RT+TMZ) in pediatric patients with newly diagnosed high-grade glioma (HGG). Methods The randomized, parallel group, multicenter, open-label HERBY trial ( ClinicalTrials.gov identifier: NCT01390948) enrolled patients age ≥ 3 years to ≤ 18 years with localized, centrally neuropathology-confirmed, nonbrainstem HGG. Eligible patients were randomly assigned to receive RT + TMZ (RT: 1.8 Gy, 5 days per week, and TMZ: 75 mg/m2 per day for 6 weeks; 4-week treatment break; then up to 12 × 28-day cycles of TMZ [cycle 1: 150 mg/m2 per day, days 1 to 5; cycles 2 to 12: 200 mg/m2 per day, days 1 to 5]) with or without BEV (10 mg/kg every 2 weeks). The primary end point was event-free survival (EFS) as assessed by a central radiology review committee that was blinded to treatment. We report findings of EFS at 12 months after the enrollment of the last patient. Results One hundred twenty-one patients were enrolled (RT+TMZ [n = 59]; BEV plus RT+TMZ [n = 62]). Central radiology review committee–assessed median EFS did not differ significantly between treatment groups (RT+TMZ, 11.8 months; 95% CI, 7.9 to 16.4 months; BEV plus RT+TMZ, 8.2 months; 95% CI, 7.8 to 12.7 months; hazard ratio, 1.44; P = .13 [stratified log-rank test]). In the overall survival analysis, the addition of BEV did not reduce the risk of death (hazard ratio, 1.23; 95% CI, 0.72 to 2.09). More patients in the BEV plus RT+TMZ group versus the RT+TMZ group experienced one or more serious adverse events (n = 35 [58%] v n = 27 [48%]), and more patients who received BEV discontinued study treatment as a result of adverse events (n = 13 [22%] v n = 3 [5%]). Conclusion Adding BEV to RT+TMZ did not improve EFS in pediatric patients with newly diagnosed HGG. Our findings were not comparable to those of previous adult trials, which highlights the importance of performing pediatric-specific studies.

Purpose To evaluate intellectual functioning and the implications of limiting radiation exposure in the four biologically distinct subgroups of medulloblastoma: wingless (WNT), sonic hedgehog (SHH), Group 3, and Group 4. Patients and Methods A total of 121 patients with medulloblastoma (n = 51, Group 4; n = 25, Group 3; n = 28, SHH; and n = 17, WNT), who were treated between 1991 and 2013 at the Hospital for Sick Children (Toronto, Ontario, Canada), Children’s National Health System (Washington, DC), or the Lucile Packard Children’s Hospital (Palo Alto, CA), had intellectual assessments. First, we compared intellectual trajectories between subgroups. Next, we evaluated the effect of treatment with reduced-dose craniospinal irradiation (CSI) plus a tumor bed boost versus treatments that deliver higher CSI doses and/or larger boost volumes to the brain (all other treatments) within subgroups. Linear mixed modeling was used to determine the stability or change in intelligence scores over time. Results Intellectual outcomes declined comparably in each subgroup except for processing speed; SHH declined less than Group 3 ( P = .04). SHH had the lowest incidence of cerebellar mutism and motor deficits. Treatment with reduced-dose CSI plus a tumor bed boost was associated with preserved intellectual functioning in WNT and Group 4 patients considered together (ie, subgroups containing patients who are candidates for therapy de-escalation), and not in Group 3 or SHH. Across all subgroups, patients in the all other treatments group declined over time (all P < .05). Conclusion SHH patients appear to have the most distinct functional (ie, motor deficits and mutism) outcomes and a unique processing speed trajectory. Only WNT and Group 4 patients seem to benefit from limiting radiation exposure. Our findings highlight the value of conducting subgroup-specific analyses, and can be used to inform novel biologically based treatment protocols for patients with medulloblastoma.

10011 Background: Responses to immune checkpoint inhibitors (ICI) in unselected pediatric solid tumors have been disappointing. We hypothesized that pediatric cancers with an increased tumor mutation burden (TMB) and/or replication repair deficiency (RRD) may be uniquely susceptible to ICI therapy. Methods: OZM-075 (NCT2992964) is a multicenter pilot study in patients (pts) aged 1-25 yrs with relapsed/refractory cancers (including CNS tumors). Pts were eligible only if tumor showed evidence of RRD (based on immunohistochemistry showing loss of expression of relevant RR protein) or elevated TMB > 5mut/Mb determined by gene panel DNA sequencing. Pts with measurable or evaluable disease were eligible. Enrolment began in May 2017 and was completed in Nov 2020. Pts received programmed death-1 (PD-1) inhibitor nivolumab (NIVO) at a fixed dose of 3mg/kg every 2 weeks. Primary objective was objective response rate (ORR) by iRECIST or iRANO for those with measurable disease. Tumor tissue and serial blood samples were collected for analysis of biomarkers of response and survival. Results: As of 26Jan2021, 6 male and 5 female patients aged 9-18 yrs received treatment with NIVO (5 glioblastoma (GBM), 2 anaplastic astrocytoma (AA), 2 neuroblastoma, 1 colorectal adenocarcinoma (CRC), 1 adrenocortical carcinoma). 3pts had TMB 5-10mut/Mb, 5pts had TMB > 10mut/Mb (range 14-837, median 64), 3 pts enrolled on basis of RRD. Time on treatment ranged from 0-21 months and 2 pts remain on therapy. Best OR was partial response (PR) in 2 pts (both GBM), with an additional patient with CRC achieving pathological CR. Remarkably one of the pts with PR showed early evidence of progression on imaging and elected to discontinue NIVO. Without further therapy, subsequent imaging 7 months later showed PR. Best response of stable disease was demonstrated in 5 further pts, lasting ≥5 months in 3 pts. Median overall survival (OS) not reached. Of 7 patients with relapsed GBM/AA, 5 were alive for ≥12 months following start of NIVO including 4 pts with confirmed TMB > 10mut/Mb. Conclusions: This pilot study showed evidence of impressive clinical benefit particularly for relapsed GBM/AA patients for whom median survival is typically ̃6 months (JCO 1995 13(1):112-). These data contrast previous pediatric data and suggest that in the context of RRD and hypermutation ICI therapy may provide profound long-term response and survival for these children. Analysis of other correlative biomarkers is ongoing. Clinical trial information: 2992964.

Purpose Recurrent glioblastoma multiforme (GBM) is incurable with current therapies. Biallelic mismatch repair deficiency (bMMRD) is a highly penetrant childhood cancer syndrome often resulting in GBM characterized by a high mutational burden. Evidence suggests that high mutation and neoantigen loads are associated with response to immune checkpoint inhibition. Patients and Methods We performed exome sequencing and neoantigen prediction on 37 bMMRD cancers and compared them with childhood and adult brain neoplasms. Neoantigen prediction bMMRD GBM was compared with responsive adult cancers from multiple tissues. Two siblings with recurrent multifocal bMMRD GBM were treated with the immune checkpoint inhibitor nivolumab. Results All malignant tumors (n = 32) were hypermutant. Although bMMRD brain tumors had the highest mutational load because of secondary polymerase mutations (mean, 17,740 ± standard deviation, 7,703), all other high-grade tumors were hypermutant (mean, 1,589 ± standard deviation, 1,043), similar to other cancers that responded favorably to immune checkpoint inhibitors. bMMRD GBM had a significantly higher mutational load than sporadic pediatric and adult gliomas and all other brain tumors (P < .001). bMMRD GBM harbored mean neoantigen loads seven to 16 times higher than those in immunoresponsive melanomas, lung cancers, or microsatellite-unstable GI cancers (P < .001). On the basis of these preclinical data, we treated two bMMRD siblings with recurrent multifocal GBM with the anti–programmed death-1 inhibitor nivolumab, which resulted in clinically significant responses and a profound radiologic response. Conclusion This report of initial and durable responses of recurrent GBM to immune checkpoint inhibition may have implications for GBM in general and other hypermutant cancers arising from primary (genetic predisposition) or secondary MMRD.

2503 Background: CNS-NGGCT are rare tumors that have been successfully treated with multimodal therapies. With a 5-yr EFS and OS of 72-84% and 82-93% respectively, surveillance and relapse detection is essential. Tumor marker (TM) elevation has proven to be a highly sensitive method of relapse detection in extra-cranial-NGGCT. We aim to determine the role of TM for relapse surveillance in children and adolescents with CNS-NGGCTs. Methods: European and North American data from germ cell tumor trials (SIOP GCT96, SFOP-TGM TC 90/92, COG-ACNS0122 and COG-ACNS1123) were pooled for analysis. Additionally, patients treated in the UK, Germany and France under strict protocol-guidelines were included. Details regarding imaging, pathology and TM elevation at diagnosis and relapse were collected. We report the proportion of relapses detectable by TM elevation. Results: Four-hundred and eighty-four patients enrolled in prospective cooperative group CNS-NGGCT trials from 1989 to 2016 were pooled for analysis. One-hundred and thirteen (23%) patients experienced a relapse/progression (SIOP GCT96: n = 57; SFOP TGM TC 90-92 n = 23, COG-ACNS0122 n = 16 & COG-ACNS1123 n = 17) and constitute the population of this report. Median age at diagnosis was 13 (range:1-30) years. The most common primary location was pineal in n = 60 (53%) patients. The site of relapse was available for 100 patients, 48 patients relapsed locally, 36 relapsed with distant disease, combined relapses were seen in 22 patients and 4 patients relapsed with TM elevation alone. TM in serum and/or CSF at diagnosis was available in 93(82%) patients, and in 90(80%) patients at the time of relapse. Eighty-four patients had TM available at both timepoints. At diagnosis 81 (96%) patients had TM elevation and 3 (4%) had negative TM. At relapse, 74(94%) patients with positive TM at diagnosis had TM elevation, while 7(6%) had TM negative. Conversely, 2/3 patients with negative TM at diagnosis, relapsed with elevated TM. Conclusions: Herein, we have assembled the largest prospective cohort to date of relapsed intracranial germ cell tumors. TM are highly sensitive detecting relapse/progression in CNS-NGGCT patients with elevated TM at diagnosis. The routine use of TM for relapse surveillance in patients with CNS-NGGCT can decrease the frequency of cross-sectional imaging, therefore, reducing lengthy hospital visits, sedation procedures and decreasing health-care costs.

10506 Background: Low-grade gliomas (LGGs) are the most common brain tumors among children. Pediatric LGGs are often not surgically resectable and tend to demonstrate relapsed/remitting courses with current standard chemotherapy regimens. Moreover, radiation is often avoided due to its associated neurocognitive and endocrinologic sequelae. However, in pediatric patients (pts) with BRAF V600-mutant LGG, dabrafenib monotherapy has demonstrated meaningful clinical activity and acceptable tolerability (Hargrave et al, Clin Cancer Res. 2019; NCT01677741). Here we report the efficacy and safety of dabrafenib + trametinib (D+T) combination therapy in pediatric pts with previously treated BRAF V600-mutant LGG. Methods: This is a 4-part, open-label, multicenter, phase I/II study (NCT02124772). The limited dose-escalation (ESC) portion evaluated the D+T combination in pediatric pts ( < 18 y) with recurrent/refractory BRAF V600-mutated solid tumors that were naive to MAPK pathway–targeted therapy. This was followed by a tumor cohort expansion (EXP), and the D+T combination was evaluated in BRAF V600-mutant LGG pts at recommended dose levels. Efficacy was determined by both investigator and independent review using the RANO criteria (for gliomas). Adverse events (AEs) were assessed per NCI-CTCAE v4.03. Results: Overall, 36 pediatric pts with LGG received D+T combination therapy (ESC, n = 16; EXP, n = 20); pooled efficacy data were available for both ESC and EXP, while LGG-specific safety data were available for EXP. At interim analysis (Aug 2019), 17 of the 20 pts in EXP remained on protocol therapy. Three pts withdrew/discontinued treatment because of AEs. Skin toxicity (95%) and pyrexia (75%) were the frequent AEs reported. No on-treatment deaths were reported. Across both ESC and EXP, the objective response rate (ORR) was 25% (95% CI, 12%–42%) per independent review (1 complete response [CR], 8 partial response [PR], 24 stable disease [SD], 2 progressive disease [PD], 1 unknown [UNK]) and 50% (95% CI, 33%–67%) per investigator review (2 CR, 16 PR, 17 SD, 1 UNK). However, ORR + SD was similar, with 92% and 97% of pts having SD or better per independent and investigator review, respectively. Conclusions: In pediatric pts with pretreated BRAF V600-mutant LGG, D+T combination therapy demonstrated clinical activity, with 92% of pts having SD or better by independent review using the RANO criteria. Pyrexia and skin toxicity were the common AEs; majority of these were low-grade and manageable. Clinical trial information: NCT02124772.

Purpose This phase II trial evaluated the effect of neoadjuvant chemotherapy with or without second-look surgery before craniospinal irradiation on response rates and survival outcomes in children with newly diagnosed nongerminomatous germ cell tumors. Patients and Methods Induction chemotherapy consisted of six cycles of carboplatin/etoposide alternating with ifosfamide/etoposide. Patients demonstrating less than complete response after induction chemotherapy were encouraged to undergo second-look surgery. Patients who did not achieve complete response or partial response after chemotherapy with or without second-look surgery proceeded to high-dose chemotherapy with thiotepa and etoposide and autologous peripheral blood stem-cell rescue before craniospinal irradiation. Results The study included 102 patients treated between January 2004 and July 2008. Median age was 12 years, and 76% were male; 53.9% had pineal region masses, and 23.5% had suprasellar lesions. Sixty-nine percent of patients achieved complete response or partial response with neoadjuvant chemotherapy. At 5 years, event-free survival was 84% ± 4% (SE) and overall survival was 93% ± 3%. During the median follow-up of 5.1 years, 16 patients recurred or progressed, with seven deaths after relapse. No deaths were attributed to therapy-related toxicity. Relapse occurred at the site of primary disease in 10 patients, at a distant site in three patients, or both in one patient. In two patients, progression was detected by marker increase alone. Increased serum α-fetoprotein was a negative prognostic variable. Histologic subtype and increase of beta-human chorionic gonadotropin were not significantly correlated with worse outcomes. Conclusion Neoadjuvant chemotherapy with or without second-look surgery achieved high response rates contributing to excellent survival outcomes in children with newly diagnosed nongerminomatous germ cell tumors. This regimen should be included as a backbone for further studies.

DOI: 10.1200/JCO.2014.59.6627 A few years ago I had the good fortune to join a research team that intended to create a device to help dying children express their wants and needs despite communication challenges. The brain tumor team at SickKids had cared for several children whose sensory and motor function had changed unpredictably, either suddenly or slowly, making it increasingly difficult to speak, write, or even to gesture to their family members. As a child life specialist on the neurosurgery team, my role was to support children’s understanding of illness and treatment; to express their emotions; to prepare for and cope with procedures, changes in their bodies, changes in their abilities, and the impacts of the disease on them and their families. This is how I met Madison, an amazing 9-year-old girl, and her family when she was diagnosed with an inoperable brain tumor. Over the next few months Madison lost her hearing; communication became louder for everyone but her, until finally, when she couldn’t hear at all, conversations were written rather than spoken. During clinic visits, I would transcribe with Crayola markers, back and forth between Madison and her care team. A position of honor. Between visits, she would e-mail her questions to me in a large font and communicated with family and friends by using a homemade messaging program that her dad had devised for her. When her vision became too blurred for her to be able to read, her family developed an intricate system of gestures and modified sign language. They learned to lip read and used exaggerated facial expressions to convey their emotions. Their intuition became so finely tuned, they seemed to “know” what each of them was thinking, and they shared their love and support with a look or a touch. Without the benefit of words, it seemed that this girl and her family had found a way to communicate what was most important at the end of her life. They were not the first or the last family to go through this, but our team was struck by how patiently they’d devised and mastered so many creative communication strategies at a time when most people don’t have the energy to be patient or creative. And of course, none of these brilliant strategies made it easy to communicate or to cope with what was happening, but they did help, and they inspired our team. We set out to create a user-friendly computer program that a child could use to scroll through categories of messages, activating the appropriate word or phrase in response to a touch, gesture, or even an eye movement. A system like that could give a child back some control, their own voice— something to counter the overwhelming feelings of helplessness. With this in mind, we set out to interview bereaved parents who had been through similar experiences to find out what this device should do, what messages to include, and what it should look like. At the time, I thought the questions we were asking were concrete and straightforward, but the responses I’d anticipated were vastly different from what was stirred up for parents. Our simple questions were met with profound responses. For instance, to populate a list of messages to include in the device, I asked, “What was important to talk about?” I started to feel awkward asking this question because it continued to be met with a puzzled expression because, obviously, “everything was important.” Making sure that children were not in pain, or cold, or hungry and that they were comfortable, happy, and ideally not bored trumped everything else for parents. Some people described tools they used to figure out where the child’s pain was, including a stuffed cat: the child would point to the part of the cat’s body where his own pain was. Tools were helpful for talking about feelings too, although this was a topic requiring care, intuition, and precision. Many families used charts of “feeling faces,” pointing to the one that showed how the child was feeling. One parent described the significant difference between “angry” and “frustrated” and the importance of validating the right feeling; although it may feel like a subtle difference to some, for a child with such limited opportunities for expression, this distinction was critical. Being able to talk about “regular stuff” was also extremely important for families. This “stuff” could be anything from gossip at school, the week’s errands, or what movie to watch. Doing what felt “normal” was described as being beneficial for the ill JOURNAL OF CLINICAL ONCOLOGY A R T O F O N C O L O G Y VOLUME 33 NUMBER 11 APRIL 1

Purpose To evaluate the efficacy of single-agent vinblastine in pediatric patients with recurrent or refractory low-grade glioma. Patients and Methods Patients were eligible if they had experienced previous treatment failure (chemotherapy and/or radiation) for incompletely resected or unresectable low-grade glioma (LGG). Vinblastine (6 mg/m2) was administered weekly for 1 year unless unacceptable toxicity or progression (confirmed on two consecutive imaging studies) occurred. Results Fifty-one patients (age range, 1.4 to 18.2 years; median age, 7.2 years) were prospectively enrolled onto this phase II study. Fifty patients had previously received at least one prior regimen of chemotherapy, and 10 patients had previously received radiation treatment. Fifty patients were evaluable for response; 18 patients (36%) had a complete, partial, or minor response, and 31 patients completed 1 year of treatment. At a median follow-up of 67 months, 23 patients had not experienced progression; three patients have died. Five-year overall survival was 93.2% ± 3.8%, and 5-year progression-free survival was 42.3% ± 7.2%. Toxicity was manageable and mostly hematologic, although a few patients needed transfusions. Conclusion Weekly vinblastine seems to be a reasonable alternative to radiation for pediatric patients with LGG who have experienced treatment failure with first-line chemotherapy. The 5-year progression-free survival observed in this phase II trial is comparable to results observed with first-line chemotherapy in chemotherapy-naive patients. The role of single-agent vinblastine and other vinca alkaloid in the management of pediatric LGGs deserves further investigation.

e22506Background: Medulloblastoma is comprised of four distinct molecular subgroups (Wnt, SHH, Group3 ,Group4) with unique clinical and prognostic implications. However, this relationship is rarely...

10506Background: The primary treatment for many pediatric patients (pts) with low-grade glioma (LGG) remains surgical resection with curative intent. However, pts whose tumors cannot be completely ...

10505Background: The BRAF V600 mutation constitutively activates the MAPK/ERK pathway and is an oncogenic driver in many tumor types, including pediatric brain cancers. There are limited effective ...

Purpose Choroid plexus carcinomas are pediatric tumors with poor survival rates and a strong, but poorly understood, association with Li-Fraumeni syndrome (LFS). Currently, with lack of biologic predictors, most children are treated with aggressive chemoradiation protocols. Patients and Methods We established a multi-institutional tissue and clinical database, which enabled the analysis of specific alterations of the TP53 tumor suppressor and its modifiers in choroid plexus tumors (CPTs). We conducted high-resolution copy-number analysis to correlate these genetic parameters with family history and outcome. Results We studied 64 patients with CPTs. All individuals with germline TP53 mutations fulfilled LFS criteria, whereas all patients not meeting these criteria harbored wild-type TP53 (P < .001). TP53 mutations were found in 50% of choroid plexus carcinomas (CPCs). Additionally, two sequence variants known to confer TP53 dysfunction, TP53 codon72 and MDM2 SNP309, coexisted in the majority of TP53 wild-type CPCs (92%) and not in TP53 mutated CPC (P = .04), which suggests a complementary mechanism of TP53 dysfunction in the absence of a TP53 mutation. High-resolution single nucleotide polymorphism (SNP) array analysis revealed extremely high total structural variation (TSV) in TP53-mutated CPC tumor genomes compared with TP53 wild-type tumors and choroid plexus papillomas (CPPs; P = .006 and .004, respectively). Moreover, high TSV was associated with significant risk of progression (P < .001). Five-year survival rates for patients with TP53-immunopositive and -immunonegative CPCs were 0% and 82 (± 9%), respectively (P < .001). Furthermore, 14 of 16 patients with TP53 wild-type CPCs are alive without having received radiation therapy. Conclusion Patients with CPC who have low tumor TSV and absence of TP53 dysfunction have a favorable prognosis and can be successfully treated without radiation therapy.

Purpose Medulloblastoma is the prototype of treatment success in modern pediatric neuro-oncology. Unfortunately, 20% to 30% of tumors recur despite maximal resection and multimodal therapy. Multiple biologic prognostic markers have been investigated to predict recurrences, but controversy remains regarding their clinical utility. Because p53 immunopositivity is an adverse prognostic marker in pediatric medulloblastoma and TP53 mutations are associated with chemotherapy and radiation therapy resistance, we aimed to determine the extent and role of TP53 mutations in pediatric medulloblastoma treatment failure. Patients and Methods One hundred eight of 111 consecutive patients diagnosed with medulloblastoma in our institution from 1995 to 2007 were included. Median follow-up time was 5.3 years in survivors. All samples were immunostained for p53 and erbB-2. Histologic grade and immunostaining were scored by two blinded reviewers. For 49 patients, frozen material was available for TP53 sequencing. The main outcome measures were overall and progression-free survival. Results Sixteen percent of sequenced medulloblastomas harbored a TP53 mutation. As a screening test, p53 immunohistochemistry was 100% sensitive and 83% specific for a TP53 mutation. Strikingly, all mutated tumors recurred early, and 5-year survival for average-risk patients was 0% for TP53-mutated medulloblastoma compared with 74% ± 8% for wild-type medulloblastoma (P < .0001). Furthermore, 75% of recurrences in average-risk patients were associated with TP53 mutations. On multivariate analysis, TP53 mutation status was the strongest adverse prognostic factor (hazard ratio = 10.4, P = .003). Conclusion Lack of long-term survival in TP53-mutated medulloblastomas highlights the role of TP53 mutations in medulloblastoma resistance to conventional therapies and the need for alternative treatments, and prospective validation of these findings is needed.

10503 Background: RAS/MAPK pathway mutations have been identified as the major drivers of pediatric low-grade glioma (pLGG). The impact of these alterations on outcome and response to therapy is still unknown. Methods: We performed a large population based study of all pLGG diagnosed from 1985-2015. Detailed treatment and very long term outcome data was collected on all patients. Known pLGG-related alterations were detected using NanoString and QX200™Droplet Digital™PCR. Molecular data was correlated with outcome and response to chemotherapy. Results: In our cohort of 614 patients, BRAF was found to be altered in 57% and wild-type (WT) in 43% of patients without neurofibromatosis 1 (NF1). Among BRAF-WT we identified H3.3K27M, FGFR1-TACC1, MYBL1 and other alterations. Molecular alterations stratified pLGG into several risk groups. Ten-year progression free survival (PFS) was 72.3% for NF1, 69.5% for KIAA1549-BRAF, 53.5% for BRAF-WT, 30.3% for BRAF-V600E and 0% for H3.3K27M mutations (p < 0.0001). Similarly, overall survival (OS) at 10 years delineated difference between excellent survival of KIAA1549-BRAF and NF1 compared to BRAF-V600E and BRAF-WT (p = 0.0005). Interestingly, all patients with FGFR1-TACC1 and MYBL1 were alive despite observed progressions. Strikingly, response to chemotherapy determined by changes in tumor size at 6 months of therapy correlated with pLGG alteration. Objective response to first line chemotherapy was observed in 46% of patients with KIAA1549-BRAF and 35% of NF1. In contrast, only 15% BRAF-V600E and 18% BRAF-WT responded and 41% tumors grew after six months of chemotherapy. Moreover, 5-year PFS after chemotherapy was strikingly low for BRAF-V600E and BRAF-WT (25% and 31.9% respectively) compared to KIAA1549-BRAF (50%) and NF1 (76.7%) (p = 0.001). This translated to decreased OS for BRAFV600E and BRAF-WT patients (p = 0.042). Conclusions: Our study provides evidence that molecular alterations dictate the outcome of pLGG. KIAA1549-BRAF harbors excellent prognosis and choice of therapy should be made in favor of less toxic agents to minimize deleterious late effects. In contrast, poor prognosis is associated with lack of response to chemotherapy in BRAF-V600E and BRAF-WT tumors.

2029 Background: Neurological side effects associated with childhood brain tumors and their treatments contribute to long term neurocognitive morbidity. The aims of this study were to identify the incidence of sensorineural hearing loss (SNHL) in a large sample of children treated for malignant brain tumors, and to evaluate the potential relationship between SNHL and intellectual functioning following the completion of treatment. Methods: We conducted a prospective follow-up study at a single center with review of 119 patients treated for embryonal brain tumors at the Hospital for Sick Children, between 1996-2015, to analyze the impact of significant SNHL (Chang > 2b) on intellectual function. Hearing was assessed post-treatment (median age: 13.5y (+4.5)) and the median age for neurocognitive testing was 12.8y (+ 4.1). The median interval from time of diagnosis was 5.8y (+ 3.7). Results: Severe SNHL was identified in half the patients (50.4%, n = 60/119). We identified a subset of patients (n = 61) who had assessments of intellectual function. In this cohort, intellectual function was significantly poorer in the group with severe SNHL, even after controlling for the effect of craniospinal radiation (severe SNHL 22.4 Gy + 13.3, no or mild hearing loss 20.4 Gy +12.8) and boost dose and volume. Children experiencing severe SNHL had lower overall IQ (severe SNHL 72.4 + 16.6; no/mild hearing loss 92.0 + 20.5) p < 0.001 and in significantly lower verbal comprehension (severe SNHL 78.7 + 15.9; no/mild hearing loss 94.7 + 13.8) p < 0.001, and working memory (severe SNHL 78.2+ 17.6; no/mild hearing loss 94.8 + 16.4) p < 0.001, scores. Conclusions: Hearing loss is a much more significant complication in children with embryonal brain tumors than previously estimated. We show the profound impact of hearing loss on intellectual deficit in children. Namely, patients with severe SNHL have difficulty using and understanding verbal language, and they have a reduced ability to concentrate and manipulate information in short-term memory. Our results have implications on future trial designs and follow-up of children treated for embryonal brain tumors.

10505 Background: Nodular desmoplastic medulloblastoma and medulloblastoma with extensive nodularity (ND/MBEN) have been associated with a more favorable outcome in younger children. However, treatment-related neurotoxicity remains a significant concern in this vulnerable group of patients. Methods: We prospectively conducted a single-arm multicenter trial of conventional chemotherapy for non-metastatic ND/MBEN, based on a modified HIT SKK2000 regimen excluding the use of intraventricular methotrexate (MTX) injection, with the aim to achieve a similar outcome with reduced treatment related neurotoxicity. The design required 37 patients and targeted a 2-year PFS of ≥ 90%. Secondary objectives included evaluation of feasibility of timely central pathology review, prospective evaluation of the cohort’s molecular profile and neurocognitive outcomes. Results: Between 12/2013 and 07/2016, 26 patients were enrolled, including 16 males and 10 females, diagnosed at a median age of 19.7 months (7.1-42.9 months). Four patients had residual disease at baseline. There were 19 ND and 7 MBEN medulloblastoma, confirmed by central pathology review. All cases were reviewed within 10 days by at least 2 of the 3 neuropathologists. The study was closed early following interim analysis due to a higher than expected relapse rate. At last follow-up, 7 patients had relapsed (3 local, 2 distant and 2 combined) at a median time of 9.7 months from diagnosis (range, 9.5-13.7 months). One patient subsequently died of disease. The current median follow-up for the 25 survivors is 1 year (range, 0.2-1.9 years) and the 1 year PFS rate is 66.2% (SE 12.2%). Based on the currently available information, older age (p = 0.07) and ND histology (p = 0.009) appear to be associated with worse PFS. To date none of the patients with MBEN histology have relapsed. Conclusions: The proposed modified regimen of chemotherapy without intraventricular MTX failed to achieve the desirable 2 y PFS of 90%, leading to premature closure of the study. Ongoing molecular characterization of the cohort may help uncover patients who may still benefit from this regimen. Clinical trial information: NCT02017964.

10509Background: Pediatric low-grade gliomas (PLGG) are the most common CNS neoplasm in children. PLGG are histologically and clinically diverse and molecular stratification is desperately needed. ...

e22003Background: Recently it has been shown that the four principle subgroups of medulloblastoma can be discerned using conventional MRI. Herein we aim to provide robust neuroimaging correlates of...

The objectives of this study were to determine (1) the role of selection before bone marrow transplantation (BMT), (2) the role of vincristine, melphalan, and total body irradiation (TBI) as consolidation of induction therapy for stage IV over 12 months at diagnosis, and (3) the role of immunomagnetic purging in metastatic neuroblastoma. Among 72 consecutive unselected patients, 10 were not grafted (four died at induction: two in complete remission [CR], two in partial remission [PR]); three had bone marrow progression before harvest; one had uncontrolled progression; and two had parental refusal). Sixty-two patients were grafted (23 in CR/very good PR [VGPR] and 39 in PR). Among the 62, 33 were consolidated with at least 90% excision of their initial tumor excised (53.2%), 15 with catecholamine secretions (24.2%), 22 with minor bone marrow involvement (35.5%), and 31 with positive bone scan (50%). Median observation time is 59 months. Progression-free survival (PFS) for the 10 excluded patients was 20% at 2 years and 0% at 4 years. PFS for the grafted population (n = 62) is 40% at 2 years, 20% at 4 years, and 13% at 7 years. No difference was observed between patients grafted in CR/VGPR or in PR. However, a group of 19 children was grafted resulting in complete normalization of metastasis (regardless of primary-site tumor status). In this group, PFS at 59 months was 38% with no relapses up to 7 years post-BMT. A group of 31 patients with no bone involvement at BMT was also identified. PFS at 5 years is 30% compared with 12% for bone-positive patients at BMT. Moreover, the 11 children presenting at diagnosis with no bone involvement (Evans stage IVS or stage C Memphis) and consolidated with BMT had PFS at 5 years of 50% with no late relapses. A subgroup of stage IV neuroblastoma patients older than 1 year of age at diagnosis may be curable with this therapeutic approach, and the use of multivariate analyses to search for prognostic factors is warranted in currently existing international registries.

e21023 Background: Children with high grade gliomas (HGG) continue to have a poor prognosis despite the use of multimodality therapy. HGG may rarely occur as part of cancer predisposition syndromes (CPS) and to be associated with other primary second cancers (PSC). We aim to evaluate the incidence of PSC with HGG at King Hussein Cancer Center, Jordan Methods: We retrospectively reviewed neurooncology database between 2003 -2013. We included patients > 3 and < 18 years and in whom we could review and confirm HGG diagnosis. We described their clinical characteristics, survival and incidence of PSC Results: 42 patients were identified (29 males, 24 Jordanians). Median age was 11.3 years (3.1-17.9). Main presenting symptoms were increased intracranial pressure in 22 (52%), focal deficits in 13 and convulsions in 10. Fourteen (67%) patients with available clinical data had cafe au lait spots (CAL). Family history of cancer was reported in 13/31 and consanguinity in 10/30. Twelve patients (29%) had near total t...

2019 Background: To date, the first line chemotherapy treatment in the majority of countries for children with NF-1 and OPG is vincristine + carboplatin. Toxicity of this regimen consists mostly in...

10004 Background: Dabrafenib is an orally available, selective ATP-competitive inhibitor of BRAF V600E kinase, approved in unresectable or metastatic melanoma pts with the V600E mutation. This international study was designed to determine the recommended phase 2 dose (RP2D) in pts 12yrs and [≤] 12yrs. Doses of 3.0, 3.75, 4.5, and 5.25mg/kg/day were assessed by a rolling six design. When a dose level was filled and under assessment, additional pts could join the previous dose level deemed tolerable. Results: Enrollment for phase 1 completed at 27 pts, 15 male, median age 9yrs (range 1-17), with HGG n = 8, LGG n = 15, LCH n = 2, and OST n = 2. One pt had a dose-limiting toxicity (DLT) of grade 3 maculopapular rash (MR) at 4.5mg/kg/day, but is on study {approx}9mths after restarting dabrafenib at 3.75mg/kg/day. Serious adverse events judged related to dabrafenib included MR (1 pt); hypotension, disseminated intravascular coagulation, fever (1 pt, outside DLT period); and arthralgia (1 pt). Duration on study ranged from 9wks to 19mo (ongoing); 20 pts remain on treatment. The RP2D for pts > 12yrs is 4.5mg/kg/day (median AUC 5285) and 5.25mg/kg/day (median AUC 4384) for [≤] 12yrs. Investigator-assessed best radiographic responses included 3 complete response (CR), 3 partial response (PR) and 2 progressive disease (PD) in HGG; 8 PR, 6 stable disease (SD) and 1 PD in LGG; 2 CR in LCH; 1 SD and 1 PD in OST (source data verification ongoing). Conclusions: The RP2D of dabrafenib for children > 12yrs and for those [≤] 12 was determined based on the target AUC when taken twice daily. The drug was well-tolerated with manageable toxicity. A high proportion of pts demonstrated radiographic responses in this phase 1 trial in different BRAF V600E-positive tumors. The disease stratified phase 2 component of the study is underway (NCT01677741). Clinical trial information: NCT01677741.

10021 Background: Treatment of children with embryonal brain tumors (EBT) includes craniospinal irradiation. There are limited data regarding the effect of radiation therapy (RT) on pulmonary function. Methods: Protocol SJMB03 enrolled patients 3 to 21 years of age with EBT. Pulmonary function tests (PFTs) [forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) by spirometry, total lung capacity (TLC) by plethysmography and diffusing capacity of the lung for carbon monoxide (DLCO)] were obtained following completion of RT, prior to each of four courses of high-dose chemotherapy (cumulative cyclophosphamide dose, 16 g/m2), and 24 months after completion of treatment (ACT). Differences between PFTs obtained following the completion of RT and 24 months ACT were compared using exact Wilcoxon signed rank tests. Results: 303 eligible patients were enrolled between June 24, 2003 and March 1, 2010, 258 of whom had at least one PFT. Median age at diagnosis - 8.9 years (range, 3.1 to 20.4 years). Median spinal RT dose - 23.4 Gy (range, 23.4 to 50.4 Gy). Median cyclophosphamide dose was 16.24 g (range, 0 to 34.38 g). 24 months ACT, DLCO was < 75% predicted in 23% (27/115 evaluated), FEV1 was < 80% predicted in 21% (32/150 evaluated), FVC was < 80% predicted in 27% (46/168 evaluated) and TLC was < 80% predicted in 18% (24/135 evaluated) of patients. DLCO was significantly decreased 24 months ACT compared to the end of RT (median difference (MD) in % predicted, - 3.00%; p = 0.035). Race and cumulative cyclophosphamide dose were not significant predictors of DLCO. DLCO was significantly higher among males (p = 0.037) than females in a model that included time point, sex, RT dose group, RT dose*time interaction and age at diagnosis. The differences in FEV1 ((MD, - 1.00%), FVC (MD, 0.00%) and TLC (MD, -2.00%) were not statistically significant. Conclusions: Among patients with EBT treated with spinal RT, DLCO was significantly decreased 24 months after completion of treatment compared to immediately post-RT. TLC was decreased 24 months ACT, suggesting that a significant minority of patients have restrictive lung disease. Continued monitoring of this cohort to five years ACT is planned. Clinical trial information: NCT00085202.

10066 Background: The disclosure of clinically significant, validated incidental and target findings to participants in genomic research is often recommended. There have been no reports on whether attitudes of parents differ if these findings emerge from an acquired pediatric cancer versus an inherited genetic condition setting. Methods: Parents in 3 large-scale projects [Canadian Pediatric Cancer Genome Consortium (CPCGC), the Finding of Rare Genes Canada Consortium (FORGE) and the Orphan Diseases: identifying Genes and Novel Therapeutics to Enhance Treatment Project (IGNITE)] were surveyed using a mailed, validated 29-item questionnaire. Two reminders were sent. Analysis was by descriptive and Chi-square statistics. Results: Response rate: 64% (n=307/480). 40% were > 50 yrs age; more than half had a grade 12 education. 86 were parents of poor risk pediatric cancer patients and 221 were parents or individuals with rare inherited conditions. Most stated a very strong or strong right to genomic research results, irrespective if from the target condition (97%) or incidental (86%). 70% wish genetic counselling pre- and post-research testing; an additional 20% were uncertain what this entails. Almost all indicated that genomic research for childhood onset conditions should occur, regardless of whether therapy existed (99%) or not (91%). A few indicated that they would not want incidental results showing an untreatable fatal condition (17%). Most want results, even if these suggest susceptibility to multiple conditions (87%) or are of uncertain health impact (84%). Most felt a right to genomic research that showed a serious condition in siblings, whether treatable (94%) or not (89%). 74% strongly support that results discovered after death of the proband be shared with family. Conclusions: Parents of children in both cancer and inherited rare conditions genomic research do not differ in indicating a strong right and desire to receive research results, even if they are of uncertain impact, of childhood onset, or after death of the proband. Clear delineation of what will or will not be offered from genomic research should be established at the time of consent.

10029 Background: Vinblastine has shown promising activity in a phase II study in children with recurrent/refractory LGG. The aim of this study was to assess the activity of vinblastine in chemotherapy naïve children. Methods: Patients < 18 years old with unresectable or progressive LGG were eligible if they had not received any previous treatment with chemotherapy or radiation. Vinblastine was administered weekly at a dose of 6 mg/m2over a period of 70 weeks. Patients who showed progression on 2 consecutive imaging studies or evidence of clinical progression were removed from treatment. Results: 54 patients (23 female) were enrolled between 2007 and 2010. Median age at inclusion was 7 years, 13 patients were < 3 years. 32 had chiasmatic/hypothalamic tumours, 6 had evidence of dissemination. 13 had neurofibromatosis type 1. Histology was pilocytic astrocytoma (25), pilomyxoid astrocytoma (4), low grade astrocytoma variant (8); 17 patients had no histological diagnosis. Treatment was well tolerated; however, only 14 patients received full dose for the duration of the study. Most common toxicity was haematological: 40 patients who experienced grade 3+ neutropenia. There were only 6 episodes of febrile neutropenia, 3 RBC transfusions and no toxic death. Best response to chemotherapy was assessed centrally by an independent radiologist: 1 CR, 10 PR, 3 MR, 28 SD, 12 PD, for a response rate of 24.5%. With a median follow-up of 2 years (9-48 months), progression-free survival at 2 years was 72.1% (95%CI: 58.1-82.2). One patient died of progression. Conclusions: Weekly vinblastine is well tolerated in paediatric LGG patients. Although the response rate appears inferior to other common LGG regimens, progression free survival at 2 years favourably compares to most currently used regimens. Supported by a grant from the Ontario Institute Cancer Research. Clinical trial information: 1000011227.

9519 Background: Hedgehog (Hh) signaling is crucial in the development and homeostasis of many human organs and tissues. Aberrant Hh signaling is involved in tumorigenesis through promotion of cell proliferation, survival, and differentiation in wide range of human cancers, including approximately 30% of MBs. LDE225 is a potent and selective inhibitor of Smo, a key positive regulator of Hh signaling. The phase I is exploring the safety and pharmacokinetics of LDE225 in pediatric patients with advanced solid tumors that are potentially dependent on Hh signaling. Preliminary data from the ongoing phase I are presented. Methods: Dose-escalation was performed according to a Bayesian design starting at 372 mg/m2 of continuous once daily oral LDE225. Safety and preliminary efficacy of LDE225 at the maximum tolerated dose will be assessed in pediatric and adult patients with recurrent MB in a phase II expansion part. Pharmacokinetic profiles were performed at Day 1 and 21. Tumor samples were analyzed for Hh pathway activation status using a 5-gene Hh signature assay. Results: Thirty-three patients (24 MB, 3 rhabdomyoscarcoma [RMS], 3 osteosarcoma, and 1 each of neuroblastoma, gliomatosis and oligoastrocytoma) with a median age of 13 years (range, 4–17 y) have enrolled. Dose-limiting toxicity of Grade 4 creatine phosphokinase elevation occurred in 1 RMS patient out of 7 patients treated at 372 mg/m2. No dose-limiting toxicity was observed at 233 and 425 mg /m2. LDE225 at 233 and 372 mg/m2 was absorbed with a median Tmax of 2 h (range, 1–24 h). Systemic exposures were comparable with adults. Two MB patients achieved a confirmed complete response (CR) at doses of 372 and 425 mg/m2. Analysis of 14 available MB tumor samples using the 5-gene Hh signature assay showed that the 2 CR patients have Hh‑activated tumor. The remaining 12 tumor samples were from patients who did not achieve response and were determined to be Hh pathway non-activated. Conclusions: LDE225 is well tolerated in pediatric patients with advanced malignancies. Preliminary data show promising efficacy in medulloblastoma patients and support the use of the 5-gene Hh signature assay as a pre-selection tool in future trials.

9524 Background: Low grade glioma represent 30-40% off all childhood brain tumors. Optimal treatment when feasible is surgical resection resulting in cure in the majority of the cases. However, the management of progressive unresectable low-grade glioma (PULGG) remains controversial. There is increasing evidence that chemotherapy can delay or even allow avoidance of radiotherapy and/or surgery in these unresectable tumors. Several protocols have been described. However, short and long term toxicity remains a major issue in this chronic disease. Within this context, we present our experience with single agent vinorelbine, a semi-synthetic vinca alkaloid. Methods:From July 2007 to January 2011, 32 patients with recurrent (10) or newly diagnosed (22) PULGG started treatment with vinorelbine 30 mg/m² on days 0, 8 and 22 for a total of 18 cycles. Tumor site was hypothalamic/chiasmatic (19), spinal cord (1), cortex (1), cerebellum (1), brainstem (4), multifocal (3) and 2 patients had gliomatosis. Four children ...

9554 Background: Medulloblastomas are malignant tumors of the posterior fossa that occur in childhood. With the advent of craniospinal radiation (CSI) treatment survival rates improved. However, CS...

10020 Background: Ependymoma is the third most common pediatric brain tumor. Despite the importance of surgical resection and pathology, its clinical course is highly unpredictable and the biological behavior is largely unknown. As part of ongoing studies to identify potential biological and therapeutic markers for the disease, we analyzed the role of telomere maintenance, which is required for cancers to sustain growth and evade senescence. Methods: We analyzed 111 primary and recurrent ependymomas that were resected at the Hospital for Sick Children in Toronto between 1986 and 2004. A tissue array was constructed. hTERT and γH2AX expression were evaluated. Thirty-one frozen samples were additionally analyzed for telomere length (TRF) and telomerase activity (TRAP assay). Results: Of the 111 samples, 40 tumors were hTERT(-) with progression-free survival (PFS) of 83±15% and 71 were hTERT(+) with PFS of 21±9% (p No significant financial relationships to disclose.

9053 Background: Carboplatin based regimens are widely used in the treatment of unresectable pediatric LGG. Carboplatin hypersensitivity reaction (CHSR) represents a main limiting factor. Objective: To analyse the incidence, characteristics, management and impact on outcome of CHSR. Methods: National retrospective review of children diagnosed with LGG between 1988 and 2004. Inclusion criteria were age No significant financial relationships to disclose.

1556 Background: Optic pathway/hypothalamic gliomas (OPHG) are predominately low-grade tumors. However, OPHGs show an unpredictable behavior and there is no recognized histologic or molecular marker anticipating this behavior. This study was performed to investigate angiogenic features as possible independent prognostic factors. Methods: We searched the databases of the Hospital for Sick Children for patients with pathologically diagnosed OPHGs between 1985 and 2002. Tumor specimens were reviewed and reconfirmed as low-grade gliomas. Those with sufficient tissue for staining were included in the study. Sections were immunostained with factor VIII (F8) and counted for microvessel density (MD). A ratio of alpha-smooth muscle actin to F8 immunostaining was calculated to give a vascular maturity index. Vascular endothelial growth factor (VEGF) and VEGF-receptor immunostaining was performed to assess angiogenic features and MIB-1 labeling index (LI) to assess proliferation. These factors were evaluated with re...

PURPOSE: To tailor postinduction therapy for stage 4 neuroblastoma in children who are older than 1 year at diagnosis according to status after induction. PATIENTS AND METHODS: From March 1987 to December 1992, 99 patients who were consecutively admitted were included in the Lyon-Marseille-Curie East of France (LMCE)3 strategy. After induction with the French Society of Pediatric Oncology NB87 regimen and surgery, patients who were in complete remission immediately proceeded to consolidation therapy with vincristine, melphalan, and fractionated total-body irradiation (VMT). All other patients underwent a postinduction strategy before VMT, either an additional megatherapy regimen or further chemotherapy with etoposide/carboplatin. RESULTS: The progression-free survival (PFS) is 29% at 7 years from diagnosis, which compares favorably with that of a similar cohort of 72 patients previously reported by our group (LMCE1; PFS of 20% at 5 years and 8% at 14 years, P = .004). In the multivariate analysis, only age younger than 3 years at diagnosis (P = .0085) and achievement of complete or very good partial remission after NB87 and surgery (P = .00024) remained significant. The PFS of the 87 patients who were included in the postinduction strategy was significantly better than that of the comparable 62 patients on the LMCE1 study (32% v 11% at 7 years; P = .005). CONCLUSION: The progressive improvements in the LMCE results over the last 10 years suggest that improvements in supportive care measures and increases in each component of this strategy (induction, postinduction, consolidation) may all contribute to increased survival rates.