Pediatric Medulloblastoma Patients Research Articles

Radiomic-and dosiomic-based clustering development for radio-induced neurotoxicity in pediatric medulloblastoma.

Texture analysis extracts many quantitative image features, offering a valuable, cost-effective, and non-invasive approach for individual medicine. Furthermore, multimodal machine learning could have a large impact for precision medicine, as texture biomarkers can underlie tissue microstructure. This study aims to investigate imaging-based biomarkers of radio-induced neurotoxicity in pediatric patients with metastatic medulloblastoma, using radiomic and dosiomic analysis. This single-center study retrospectively enrolled children diagnosed with metastatic medulloblastoma (MB) and treated with hyperfractionated craniospinal irradiation (CSI). Histological confirmation of medulloblastoma and baseline follow-up magnetic resonance imaging (MRI) were mandatory. Treatment involved helical tomotherapy (HT) delivering a dose of 39 Gray (Gy) to brain and spinal axis and a posterior fossa boost up to 60Gy. Clinical outcomes, such as local and distant brain control and neurotoxicity, were recorded. Radiomic and dosiomic features were extracted from tumor regions on T1, T2, FLAIR (fluid-attenuated inversion recovery) MRI-maps, and radiotherapy dose distribution. Different machine learning feature selection and reduction approaches were performed for supervised and unsupervised clustering. Forty-eight metastatic medulloblastoma patients (29 males and 19 females) with a mean age of 12 ± 6years were enrolled. For each patient, 332 features were extracted. Greater level of abstraction of input data by combining selection of most performing features and dimensionality reduction returns the best performance. The resulting one-component radiomic signature yielded an accuracy of 0.73 with sensitivity, specificity, and precision of 0.83, 0.64, and 0.68, respectively. Machine learning radiomic-dosiomic approach effectively stratified pediatric medulloblastoma patients who experienced radio-induced neurotoxicity. Strategy needs further validation in external dataset for its potential clinical use in ab initio management paradigms of medulloblastoma.

Pediatric neurosurgical medulloblastoma outcomes in La Paz, Bolivia: How a Lower Middle-Income Country (LMIC) institution in South America compares to the United States.

How pediatric medulloblastoma patients fare in Lower Middle-Income Country (LMICs) in South America is not well understood. Correspondingly, the aim of this study was to summarize the pediatric neurosurgical experience of an institution in La Paz, and compare outcomes to that of a generalized High Income Country (HIC) United States (US) experience. A retrospective review of all pediatric neurosurgical medulloblastoma patients at the Children's Hospital of La Paz, Bolivia (Hospital del Niño "Dr. Ovidio Aliaga Uria") between 2014 and 2023 was conducted and compared to a generalized US experience abstracted from the US National Cancer Database (NCDB) and National Inpatient Sample (NIS) databases. Categorical, continuous and survival data were statistically summarized and compared. A total of 24 pediatric medulloblastoma patients underwent neurosurgical treatment at the Hospital del Niño. In this La Paz cohort, there were 15 (63%) males and 9 (38%) females, with a mean age of 5.6 years old at diagnosis. The majority of patients underwent subtotal resection (STR, 79%), while the remaining patients underwent biopsy only. Ten (42%) patients expired during their hospitalization, and mean length of stay overall was 39 days. Only 8 (33%) patients received adjuvant treatment after surgery. Median overall survival from diagnosis in the La Paz cohort was 1.9 months. Compared to the US databases, the La Paz cohort experienced significantly more emergency room admissions for surgery, less gross total resection, more STR, more return to operating room for ventriculoperitoneal shunting, more bacteremia, more tracheostomy procedures, more percutaneous gastrostomy placements, longer lengths of stay, less adjuvant chemotherapy, less radiation therapy, shorter follow-up, and ultimately, significantly shorter overall survival (all P < 0.050). Pediatric neurosurgical medulloblastoma outcomes at the Children's Hospital of La Paz, Bolivia are significantly inferior to that of a generalized US experience. Future research is required to identify institution- and country-specific initiatives to improve discrepancies between institutions in LMICs in South America compared to HICs.

755 Pathology in the Digital Era of Neuro-oncology: A Pathomics-Based Approach to Modeling Outcomes in Medulloblastoma

INTRODUCTION: Pathomics is an emerging data science technique that may be used to extract intricate, sub-visual histopathologic features using high-throughput digital image analysis. METHODS: Using the Children’s Brain Tumor Network database, pediatric medulloblastoma patients were evaluated for inclusion in the study. After tumor segmentation, 49 quantitative pathomic features were extracted using a digital pathology software package (Qupath). LASSO cox proportional hazard model with stratified 5-fold cross-validation was used to identify high-performing features and create a predictive survival model by identifying low-, medium- and high-risk OS cohorts. A classification model was developed using histogram gradient boosting classifier to perform binary classification of Group3/4 vs. SHH/WNT with leave-one-subject-out cross validation. RESULTS: 84 patients with median age at diagnosis of 8.61 years (range 0.31-21.72), and median OS of 45.6 months (range 0.76-195.87) were included in the study. A survival model built using only clinical features yielded a concordance index (c-index) of 0.76 compared to actual outcomes (stratification: p &lt; 0.001). However, a survival model built by combining clinical and high-performing pathomic features yielded the best performance, showing disparate outcomes between low-, medium- and high-risk groups (p &lt; 0.001) with resultant c-index of 0.85. Furthermore, pathomic features classified subjects into Group 3/4 or SHH/WNT with classification accuracy of 71% (SEM = 0.06), and t-test against stratified chance accuracy (57%) showed significantly above-chance performance (p = .019). CONCLUSIONS: This study represents the largest pathomic-based machine learning analysis for predicting survival and molecular subtypes in medulloblastoma. The results demonstrate success in pathomic analysis using machine learning-based modeling to predict survival and molecular subtypes for patients with medulloblastoma. Further work will aim to perform our analysis on an external validation dataset to enhance model performance and improve its clinical applicability.

Single-fraction Radiation Treatment Dose Response in a Genetically Engineered Mouse Model of Medulloblastoma.

Medulloblastoma is the most common malignant brain tumor of children. Although standard of care radiotherapy for pediatric medulloblastoma (PM) can lead to long-term remission or cure in many patients, it can also cause life-long cognitive impairment and other adverse effects. The pathophysiological mechanisms involved in radiation-induced cerebral damage are incompletely understood, and their elucidation may lead to interventions that mitigate radiation toxicity. To explore the mechanisms of radiation-induced cerebral damage, transgenic mouse models of PM and non-tumor-bearing controls were exposed to radiation doses that ranged from 0 to 30 Gy. Between 0-20 Gy, a significant dose-dependent reduction in tumor-associated hydrocephalus and increase in overall survival were observed. However, at 30 Gy, hydrocephalus incidence increased and median overall survival fell to near-untreated levels. Immunohistochemistry revealed that both tumor-bearing and non-tumor-bearing mice treated with 30 Gy of radiation had significantly more reactive astrocytes and microvascular damage compared to untreated controls. This effect was persistent across mice that were given 1 and 2 weeks of recovery time after irradiation. Our data suggest that radiation therapy promotes neural death by inducing long-term neuroinflammation in PM, suggesting radiation delivery methods that limit inflammation may be effective at widening the therapeutic window of radiation therapy in PM patients.

BIOS-05. USING DATA-DRIVEN ANALYTICS TO PREDICT SURVIVAL IN PEDIATRIC PATIENTS WITH MEDULLOBLASTOMA: THE UTILITY OF MACHINE LEARNING IN NEXT-GENERATION NEURO-ONCOLOGY PROGNOSTICATION

Abstract Medulloblastoma is the most common malignant intracranial tumor affecting the pediatric population. Despite advancements in multimodal treatment over the past 2 decades yielding a &amp;gt;75% 5-year survival rate, children who survive often have substantial neurological and cognitive sequelae. We aimed to identify risk factors and develop a clinically friendly online calculator for prognostic estimation in pediatric medulloblastoma patients. Pediatric patients with a histopathologically confirmed diagnosis of medulloblastoma were extracted from the Surveillance Epidemiology and End RESULTS: database (2000-2019) and split into training and validation cohorts in an 80:20 ratio (Figure 1). The Cox proportional hazards model was used to identify the univariate and multivariate survival predictors. Subsequently, a calculator with those factors was developed to predict 2-, 5-, and 10-year overall survival as well as median survival months for pediatric medulloblastoma patients. The performance of the calculator was determined by discrimination, calibration, and decision curve analysis (DCA). A total of 1,739 pediatric patients with medulloblastoma met the prespecified inclusion criteria. Fourteen variables, including age, sex, race, ethnicity, median household income, county attribute, laterality, histology, anatomical location, tumor grade, tumor size, surgery status, radiotherapy, and chemotherapy, were included in the calculator (). The concordance index was 0.757 in the training cohort and 0.762 in the validation cohort, denoting clinically useful predictive accuracy. Good agreement between the predicted and observed outcomes was demonstrated by the calibration plots. The DCA curves indicated that the developed model has a good clinical prognostic benefit for pediatric medulloblastoma patients, and could aid prognostication efforts in the clinical setting. An easy-to-use prognostic calculator for a large cohort of pediatric patients with medulloblastoma was established. Future efforts should focus on improving granularity of population-based registries and externally validating the proposed calculator.

An Innovative Approach to Scalp Segmentation in Pediatric Brain Radiotherapy: An Atlas-Based Study.

This study aims to develop an atlas-based method for segmentation of the scalp in pediatric patients that received brain radiotherapy. The study focuses on the development of a model that utilizes CT scans of pediatric medulloblastoma (MB) patients who underwent craniospinal-irradiation (CSI) and a boost in the posterior cranial fossa (PCF). A team of experts created the Alopecia_Hope (A_H) contouring protocol on the CT scan using the 2018 EPTN consensus-based atlas and by defining for the manual adjustments the anatomic bone landmarks to differentiate between the scalp and the facial and neck skin. Using RayStation version 12A-SP1, the system was running his latest available software version at the time of study. Automated scripts were created to identify bone tissue and skin as a subtraction volume between bone and body. The soft tissue CT window was then manually used to increase or decrease the thickness of the skin, following the landmarks outlined in the protocol. The entire head scalp was defined as WScalp (WS), and a duplicate and reduced version was also defined as PCFScalp (PCFS). We randomly chose 40 patients to create both WS and PCF atlas with RayStation and 8 patients were used as testers for both volumes, the patients with major bone changes after neurosurgery were excluded. The contouring time was recorded, the mean time was calculated, and the contouring results were analyzed by experts. Standard similarity metrics of 3D DICE similarity coefficient and mean distance to agreement (MDA) were calculated. A total of 52 patients were selected for the study, with a median contouring time of 19 minutes using the guidelines, the A_H protocol and automatic scripting. Three patients were excluded. Of the 8 testers, including both volumes (WS and PCFS), the mean contouring time for the expert operator was 17.5 min. While the contouring undergone by the automatic atlas was 11 min and the mean time of the expert correction was 4.5 min. Considering the work done by the atlas is in background, this resulted in a time savings of 80% for the operator. Table 1 contains mean and one standard deviation data for the metric, both WS and PCFS tested. Atlas-based self-segmentation and the Alopecia_Hope protocol offer a more efficient and accurate method to identify specific areas of the scalp that may be affected by post-radiotherapy hair loss. This study demonstrates the effectiveness and efficiency in reducing radiotherapy workload and improving accuracy, with possible future application in the prevention of permanent alopecia and in improving the quality of life of patients undergoing whole brain radiotherapy.

Barriers to Care for Pediatric Rural Medulloblastoma Patients Receiving Craniospinal Proton Radiotherapy without In-State Facilities

Medulloblastoma represents ∼20% of all pediatric brain malignancies. The preferred treatment modality for craniospinal irradiation (CSI) has evolved to be proton beam therapy to reduce late toxicities including secondary malignancy. Access to proton therapy centers is more limited in states without one. Barriers to access for rural patients are underreported in the literature. Patients <25 years old with a diagnosis of medulloblastoma were identified from a tumor registry at a rural academic center. A chart review was conducted to identify specific barriers to proton beam CSI. Descriptive analyses were performed to describe this cohort and their barriers. We identified a total of 18 patients diagnosed between 2000 and 2022. The mean age was 10 years old (range 11 months - 20 years). 2 patients were excluded who had minimal documentation. 7 patients were diagnosed before 1/1/2014, a cutoff after which protons were considered more often in this patient population. Patient demographics were 39% female, 61% male, 22% Black, and 72% White, 6% Hispanic. 3 patients received protons. 2 patients had recurrence and 3 patients developed secondary malignancy. Out of 18 patients, proton therapy was discussed with 11 patients. The rates of documented barriers to radiotherapy for the cohort at large (n = 18) and for those where proton therapy was discussed in particular (n = 11) are as follows (cohort at large / had discussion): 4(22%) / 4(36%) difficulty obtaining insurance approval, 4(22%) / 4(36%) inpatient medical needs, 5(28%) / 5(46%) outpatient medical needs, 4(22%) / 4(36%) family scheduling conflicts, 3(17%) / 3(27%) travel costs/financial burdens, 4(22%) / 4(36%) not a clinical trial candidate, 3(17%) / 3(27%) radiotherapy delay, 9(50%) / 5(46%) other barriers to non-radiotherapy care. For the cohort at large or for those with proton therapy discussions, there were no significant associations between white and underrepresented minorities for the aforementioned barriers to care. To our knowledge, this is the first study that identifies patient barriers to accessing proton beam CSI for medulloblastoma patients from a rural tertiary care center without in-state proton radiotherapy centers. This study yields insight into the particular barriers encountered by these patients and their families, allowing clinical teams to identify potential issues in an effort to overcome those barriers. Advocacy for access to care on behalf of this vulnerable patient population may be required by our field on a state and national level.

Intraventricular SHH inhibition proves efficient in SHH medulloblastoma mouse model and prevents systemic side effects.

Medulloblastoma (MB) is the most common malignant brain tumor in children and requires intensive multimodal therapy. Long-term survival is still dissatisfying and, most importantly, survivors frequently suffer from severe treatment-associated morbidities. The sonic hedgehog pathway (SHH) in SHH MB provides a promising target for specific therapeutic agents. The small molecule Vismodegib allosterically inhibits SMO, the main upstream activator of SHH. Vismodegib has proven effective in the treatment of MB in mice and in clinical studies. However, due to irreversible premature epiphyseal growth plate fusions after systemic application to infant mice and children, its implementation to pediatric patients has been limited. Intraventricular Vismodegib application might provide a promising novel treatment strategy for pediatric medulloblastoma patients. Infant medulloblastoma-bearing Math1-cre::Ptch1Fl/Fl mice were treated with intraventricular Vismodegib in order to evaluate efficacy on tumor growth and systemic side effects. We show that intraventricular Vismodegib treatment of Math1-cre::Ptch1Fl/Fl mice leads to complete or partial tumor remission only 2 days after completed treatment. Intraventricular treatment also significantly improved symptom-free survival in a dose-dependent manner. At the same time, intraventricular application prevented systemic side effects in the form of anatomical or histological bone deformities. We conclude that intraventricular application of a SHH pathway inhibitor combines the advantages of a specific treatment agent with precise drug delivery and might evolve as a promising new way of targeted treatment for SHH MB patients.

SDPS-40 MEDULLOBLASTOMA MOLECULAR SUBTYPE PREDICTION USING PATHOMIC FEATURE ANALYSIS

Abstract The clinical course of pediatric medulloblastoma patients exhibits wide variability based on molecular subtype. The ability to predict medulloblastoma molecular subtypes (Group 3/4, WNT, SHH) based on histopathologic features alone has immense utility by providing clinicians insight into the molecular and biological underpinnings of the tumor. Although the use of computational pathology (“pathomics”) is expanding in multiple disciplines, its application in pediatric neuro-oncology has been limited. Our preliminary analysis highlights the potential for application in medulloblastoma and pediatric brain tumor research. Using the Children’s Brain Tumor Network database, pediatric patients with available clinical data, whole-slide images (WSI) and molecular subtyping were evaluated for inclusion. Regions-of-interest (ROI) on each H&amp;E-stained WSI were selected for pathomic analysis, and features were extracted using a digital pathology software package (Qupath). 25 pathomic features capturing cellular and nuclear-level intensity and morphology properties were included. Subsequently, a histogram gradient boosting classifier was used to perform binary classification of Group 3/4 vs. SHH/WNT with leave-one-subject-out cross validation. A total of 68 subjects, with median age at diagnosis of 8.19 years (range 0.29-18.69), and median overall survival (OS) of 56 months (range 0.76-258.37) were identified. Our cohort included 44 and 24 subjects in Group 3/4 and SHH/WNT, respectively. Pathomic features classified subjects into Group 3/4 or SHH/WNT with a classification accuracy of 71% (SEM=0.06). A t-test against stratified chance accuracy (57%) showed significantly above-chance testing performance (p=.019). Inspection of the confusion matrix indicated better discrimination of Group 3/4 compared to SHH/WNT subjects, likely due to the larger number of these subjects in the dataset. Our preliminary analysis shows acceptable testing accuracy in distinguishing Group 3/4 from SHH/WNT medulloblastoma, and highlights the possibility of gaining molecular insight purely from histopathologic images. The potential of pathomic features should be considered an untapped resource.

MDB-11. AN ONLINE CALCULATOR USING MACHINE LEARNING FOR PREDICTION OF SURVIVAL IN PEDIATRIC PATIENTS WITH MEDULLOBLASTOMA

Abstract OBJECTIVE Medulloblastoma is the most common malignant intracranial tumor affecting the pediatric population. Despite advancements in multimodal treatment over the past 2 decades yielding a &amp;gt;75% 5-year survival rate, children who survive often have substantial neurological and cognitive sequelae. The authors aimed to identify risk factors and develop a clinically friendly online calculator for prognostic estimation in pediatric medulloblastoma patients. METHODS Pediatric patients with a histopathologically confirmed medulloblastoma were extracted from the Surveillance Epidemiology and End Results database (2000-2019) and split into training and validation cohorts in an 80:20 ratio. The Cox proportional hazards model was used to identify the univariate and multivariate survival predictors. Subsequently, a calculator with those factors was developed to predict 2-, 5-, and 10-year overall survival as well as median survival months for pediatric medulloblastoma patients. The performance of the calculator was determined by discrimination, calibration, and decision curve analysis (DCA). RESULTS A total of 1,739 pediatric patients with medulloblastoma met the prespecified inclusion criteria. Fourteen variables, including age, sex, race, ethnicity, median household income, county attribute, laterality, histology, anatomical location, tumor grade, tumor size, surgery status, radiotherapy, and chemotherapy, were included in the calculator (https://spine.shinyapps.io/Peds_medullo/). The concordance index was 0.757 in the training cohort and 0.762 in the validation cohort, denoting clinically useful predictive accuracy. Good agreement between the predicted and observed outcomes was demonstrated by the calibration plots. The DCA curves indicated that the developed model has a good clinical prognostic benefit for pediatric medulloblastoma patients. CONCLUSION An easy-to-use prognostic calculator for a large cohort of pediatric patients with medulloblastoma was established. Future efforts should focus on improving granularity of population-based registries and externally validating the proposed calculator.

In Vivo Radiobiological Investigations with the TOP-IMPLART Proton Beam on a Medulloblastoma Mouse Model.

Protons are now increasingly used to treat pediatric medulloblastoma (MB) patients. We designed and characterized a setup to deliver proton beams for in vivo radiobiology experiments at a TOP-IMPLART facility, a prototype of a proton-therapy linear accelerator developed at the ENEA Frascati Research Center, with the goal of assessing the feasibility of TOP-IMPLART for small animal proton therapy research. Mice bearing Sonic-Hedgehog (Shh)-dependent MB in the flank were irradiated with protons to test whether irradiation could be restricted to a specific depth in the tumor tissue and to compare apoptosis induced by the same dose of protons or photons. In addition, the brains of neonatal mice at postnatal day 5 (P5), representing a very small target, were irradiated with 6 Gy of protons with two different collimated Spread-Out Bragg Peaks (SOBPs). Apoptosis was visualized by immunohistochemistry for the apoptotic marker caspase-3-activated, and quantified by Western blot. Our findings proved that protons could be delivered to the upper part while sparing the deepest part of MB. In addition, a comparison of the effectiveness of protons and photons revealed a very similar increase in the expression of cleaved caspase-3. Finally, by using a very small target, the brain of P5-neonatal mice, we demonstrated that the proton irradiation field reached the desired depth in brain tissue. Using the TOP-IMPLART accelerator we established setup and procedures for proton irradiation, suitable for translational preclinical studies. This is the first example of in vivo experiments performed with a "full-linac" proton-therapy accelerator.

Clinical Outcomes of Medulloblastoma Patients Treated with Proton Radiotherapy: A Systematic Review

<h3>Purpose/Objective(s)</h3> The aim of this study was to comprehensively review all studies examining clinical outcomes of proton therapy for craniospinal irradiation of medulloblastoma, to determine whether theoretical dosimetric advantages of proton therapy have translated into superior clinical outcomes, including survival and toxicities. <h3>Materials/Methods</h3> We performed a systematic review based on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines, using the PubMed (Medline), EMBASE, and Cochrane Library databases (inception to December 2021). The study was registered with PROSPERO, CRD42022302455. Articles in the English language reporting on clinical outcomes of pediatric and/or adult medulloblastoma patients treated with proton therapy were included. Two independent authors screened all titles/abstracts, reviewed full texts and extracted relevant data, with discrepancies resolved by consensus. Individual study quality was scored using Newcastle Ottawa Scale and overall evidence base by GRADE scale. <h3>Results</h3> Thirty-five studies were included, representing 629 unique patients. Publication dates ranged from 2011 to 2021, with the majority (n=32) from the United States. None of the studies were randomized, 12 were comparative, 9 were prospective, 3 mixed and 22 retrospective. Average mean/median follow up was 5.0 years (range = 4 weeks to 12.6 years). The majority of studies (n=19) reported on treatment with passive scatter proton beams exclusively. Average study quality score was 6.08 out of 9 (SD 0.688, median 5). Seven studies scored ≥8/9 on the Newcastle Ottawa Scale; an overall "moderate" GRADE score was assigned. Well-designed comparative cohort studies with adequate follow-up demonstrate superior neurocognitive outcomes, lower incidence of hypothyroidism (23% vs 69%) and reduced acute toxicities (including myelosuppression, esophagitis, diarrhea, weight loss and nausea/vomiting) in patients treated with protons compared to photons. The 10-year cumulative incidence of secondary malignancy is also lower for proton cohorts (2.1-4.9% vs 8%), but did not reach statistical significance in any individual study. Overall survival (10-yr) (85.3-86.9% for standard-risk disease), progression-free survival (10-yr), patterns of failure, and other endocrine outcomes such as incidence of GH deficiency, adrenal insufficiency and precocious puberty were similar to those reported for photon therapy. Quality of life, ototoxicity and incidence of CNS injury were also found to be comparable between proton and photon cohorts, albeit with lower quality of evidence. Overall, quality of evidence was most robust in the domains of intellectual outcomes, endocrinopathies and acute toxicity. <h3>Conclusion</h3> Moderate grade evidence supports proton beam therapy as a preferred treatment for craniospinal irradiation of medulloblastoma based on equivalent disease outcomes and comparable-to-improved toxicity versus photon beam radiotherapy.

MEDB-65. Molecular subclassification of a national cohort of pediatric medulloblastoma based on methylation profile

INTRODUCTION: Pediatric Medulloblastoma (MB) accounts for approximately 20% of all childhood brain tumors. Molecular subgroups namely WNT, SHH, Group 3 and Group 4, exhibit divergent biology, and clinical outcomes. DNA methylation analysis is a robust option to classify pediatric MB into molecular subgroups, which allows the optimization of diagnosis and stratification of the treatment. We review the first experience of molecular subclassification carried out at the national level in our country. METHODS: Multi-center centralized prospective and retrospective study of frozen tumor samples at diagnosis from pediatric MB patients diagnosed in Spanish hospitals, from April 2021 to December 2021. A registry was created with histology review, immunohistochemical (IHC) subgrouping, and a molecular subgrouping based on the Minimal Methylation Classifier (MIMIC) from Schwalbe et al., 2017. The time from the sample centralized reception to the study result was also collected. RESULTS: 25 frozen MB tumor samples from patients at diagnosis were included. 6 were retrospective and 19 prospective. IHC classified 19 cases (76%) as non-WNT/non-SHH MBs, 3 (12%) as WNT-activated and 3 (12%) as SHH-activated. MIMIC classified, in the non-WNT/non-SHH, 6 tumors (24%) as Group 3 and 13 (52%) as Group 4. 2 cases (8%) were WNT-activated MBs and 3 (12%) were SHH-activated MBs. Only 1 case (4%) was unclassified by MIMIC (WNT using IHC). Comparing both methods (IHC and MIMIC), diagnosis agreed in 96% of cases. The response time ranged from 5 to 10 days. CONCLUSIONS: DNA methylation profiling has proven to be a robust and quick option to classify MB into subgroups and it correlates with the IHC diagnosis. This tool was successfully implemented in our national routine diagnosis, enabling a reliable and rapid molecular subgrouping classification.

MEDB-26. Outcomes of children with standard-risk and high-risk medulloblastoma treated with pre-irradiation chemotherapy and risk-adapted craniospinal irradiation: a report on patients from the Polish Pediatric Neuro-oncology Group

BACKGROUND: The last two decades have witnessed several efforts to minimize the adverse sequelae of craniospinal irradiation (CSI), a standard of care treatment modality in medulloblastoma. This has been accomplished by adding chemotherapy to the treatment backbone. The use of pre-irradiation chemotherapy has also been previously reported. In one of the largest studies to date, we analyze treatment outcomes in children with standard and high-risk medulloblastoma treated with pre-irradiation chemotherapy followed by reduced-dose radiotherapy in SR and maintenance chemotherapy. METHODS: Data from the Polish Pediatric Neuro-oncology Group (PPNG) was analyzed in patients greater than 3 years of age with newly-diagnosed medulloblastoma. RESULTS : Among 138 patients, median age at diagnosis was 7.9 years and median follow-up was 5.5 years. Comprehensive molecular subgrouping was not available for all patients at the time of data collection. Of 60 standard-risk patients, there was pre-irradiation disease recurrence in one patient. One patient expired prior to radiation due to metastatic disease. Of 78 high-risk patients, one had pre-irradiation recurrence. Overall survival (OS) for high-risk patients at 3 and 5 years (± standard error) was 89.2 ± 4.0% and 81.3 ± 5.8%, respectively. OS for standard-risk patients at 3 and 5 years was 92.5 ± 3.8% and 88.2 ± 5.1%, respectively. Among high-risk patients, event-free survival (EFS) at 3 and 5 years was 82.5 ± 5.3% and 81.0 ± 5.6%. Among standard-risk patients, 3-year EFS was 89.2 ± 4.6% and 5-year EFS was 86.8 ± 5.3%. CONCLUSION : This study demonstrates promising survival outcomes in pediatric medulloblastoma patients treated with pre-irradiation chemotherapy followed by reduced-dose CSI and adjuvant chemotherapy. Such an approach may be helpful if delays in starting radiotherapy are expected, which is usually the case in many institutions around the globe.

A New Risk Stratification System Based on MRI Features and Machine Learning to Predict Differential Prognosis for Pediatric Medulloblastoma Patients

A New Risk Stratification System Based on MRI Features and Machine Learning to Predict Differential Prognosis for Pediatric Medulloblastoma Patients

Identification of an Immunogenic Medulloblastoma-Specific Fusion Involving EPC2 and GULP1

Simple SummaryMedulloblastoma is the most common malignant childhood brain tumor and it is considered poor immunogenic because of its low mutational burden. Nevertheless, several clinical trials are currently evaluating immunotherapy for medulloblastoma patients, since new treatment strategies for this entity are a matter of utmost urgency. Tumor specific antigens resulting from gene fusions are potentially highly immunogenic. In our study, we identified a new medulloblastoma-specific fusion transcript EPC2-GULP1.The resulting protein sequence produced a neoantigen, which was able to activate CD8+ T cells. Thus, our data indicate an immunotherapeutic approach for pediatric medulloblastoma patients carrying the EPC2-GULP1 fusion or other fusions generating immunogenic neoantigens.Medulloblastoma is the most common malignant brain tumor in children. Immunotherapy is yet to demonstrate dramatic results in medulloblastoma, one reason being the low rate of mutations creating new antigens in this entity. In tumors with low mutational burden, gene fusions may represent a source of tumor-specific neoantigens. Here, we reviewed the landscape of fusions in medulloblastoma and analyzed their predicted immunogenicity. Furthermore, we described a new in-frame fusion protein identified by RNA-Seq. The fusion involved two genes on chromosome 2 coding for the enhancer of polycomb homolog 2 (EPC2) and GULP PTB domain containing engulfment adaptor 1 (GULP1) respectively. By qRT-PCR analysis, the fusion was detected in 3 out of 11 medulloblastoma samples, whereby 2 samples were from the same patients obtained at 2 different time points (initial diagnosis and relapse), but not in other pediatric brain tumor entities. Cloning of the full-length sequence indicated that the fusion protein contains the N-terminal enhancer of polycomb-like domain A (EPcA) of EPC2 and the coiled-coil domain of GULP1. In silico analyses predicted binding of the neoantigen-derived peptide to HLA-A*0201. A total of 50% of the fusions described in the literature were also predicted to produce an immunogenic peptide. The EPC2-GULP1 fusion peptide was able to induce a de novo T cell response characterized by interferon gamma release of CD8+ cytotoxic T cells in vitro. While the functional relevance of this fusion in medulloblastoma biology remains to be clarified, our data support an immunotherapeutic approach for pediatric medulloblastoma patients carrying the EPC2-GULP1 fusion and other immunogenic fusions.

Decade-long disease, secondary malignancy, and brainstem injury outcomes in pediatric and young adult medulloblastoma patients treated with proton radiotherapy.

Survivors of pediatric medulloblastoma experience long-term morbidity associated with the toxic effects of postoperative radiotherapy (RT). Proton RT limits radiation dose to normal tissues thereby reducing side effects of treatment while maintaining high cure rates. However, long-term data on disease outcomes and long-term effects of proton RT remain limited. One hundred seventy-eight pediatric medulloblastoma patients treated with proton RT between 2002 and 2016 at the Massachusetts General Hospital comprise the cohort of patients who were treated with surgery, radiation therapy, and chemotherapy. We evaluated event-free survival (EFS), overall survival (OS), and local control using the Kaplan-Meier method. The cumulative incidence of brainstem injury and secondary malignancies was assessed. Median follow-up was 9.3 years. One hundred fifty-nine patients (89.3%) underwent a gross total resection (GTR). The 10-year OS for the entire cohort, standard-risk (SR), and intermediate/high-risk (IR/HR) patients was 79.3%, 86.9%, and 68.9%, respectively. The 10-year EFS for the entire cohort, SR, and IR/HR cohorts was 73.8%, 79.5%, and 66.2%. The 10-year EFS and OS for patients with GTR/NTR were 75.3% and 81.0% vs 57.7% and 61.0% for subtotal resection (STR). On univariate analysis, IR/HR status was associated with inferior EFS, while both anaplastic histology and IR/HR status were associated with worse OS. The 10-year cumulative incidence of secondary tumors and brainstem injury was 5.6% and 2.1%, respectively. In this cohort study of pediatric medulloblastoma, proton RT was effective, and disease outcomes were comparable to historically treated photon cohorts. The incidence of secondary malignancies and brainstem injury was low in this cohort with mature follow-up.

MRI Correlates of Ototoxicity in the Auditory Pathway in Children Treated for Medulloblastoma.

To assess diffusion and perfusion changes of the auditory pathway in pediatric medulloblastoma patients exposed to ototoxic therapies. Retrospective cohort study. A single academic tertiary children's hospital. Twenty pediatric medulloblastoma patients (13 men; mean age 12.0 ± 4.8 yr) treated with platinum-based chemotherapy with or without radiation and 18 age-and-sex matched controls were included. Ototoxicity scores were determined using Chang Ototoxicity Grading Scale. Three Tesla magnetic resonance was used for diffusion tensor and arterial spin labeling perfusion imaging. Quantitative diffusion tensor metrics were extracted from the Heschl's gyrus, auditory radiation, and inferior colliculus. Arterial spin labeling perfusion of the Heschl's gyrus was also examined. Nine patients had clinically significant hearing loss, or Chang grades more than or equal to 2a; 11 patients had mild/no hearing loss, or Chang grades less than 2a. The clinically significant hearing loss group showed reduced mean diffusivity in the Heschl's gyrus (p = 0.018) and auditory radiation (p = 0.037), and decreased perfusion in the Heschl's gyrus (p = 0.001). Mild/no hearing loss group showed reduced mean diffusivity (p = 0.036) in Heschl's gyrus only, with a decrease in perfusion (p = 0.008). There were no differences between groups in the inferior colliculus. There was no difference in fractional anisotropy between patients exposed to ototoxic therapies and controls. Patients exposed to ototoxic therapies demonstrated microstructural and physiological alteration of the auditory pathway. The present study shows proof-of-concept use of diffusion tensor imaging to gauge ototoxicity along the auditory pathway. Future larger cohort studies are needed to assess significance of changes in diffusion tensor imaging longitudinally, and the relationship between these changes and hearing loss severity and longitudinal changes of the developing auditory white matter.

Gene-Specific DNA Methylation Profiles in Pediatric Medulloblastomas.

Medulloblastoma is the most common pediatric central nervous tumor of high malignancy that has been classified into both histological subtypes and molecular subgroups by the 2016 World Health Organization classification. However, there is a still need to understand the genomic characteristics and predict the clinical course. The aim of the study is to investigate the significance of the methylation profiles in molecular subclassification and precision medicine of the disease. The study enrolled 47 pediatric medulloblastoma patients. DNA methylation levels of KLF4, SPINT2, RASSF1A, EZH2, ZIC2, and PTCH1 genes were analyzed using methylation-specific pyrosequencing. The significance of the statistical relationship between methylation profiles and clinicopathological parameters including molecular subgroups and histological subtypes, the status of metastasis, and event-free survival were analyzed. DNA methylation analysis demonstrated that KLF4, PTCH1, and ZIC2 hypermethylation were associated with the SHH-activated subgroup, whereas both SPINT2 and RASSF1A hypermethylation were associated with metastatic disease. EZH2 gene was not methylated in any of the samples. We think that customized DNA methylation profiling may be a useful tool in the molecular subclassification of pediatric medulloblastoma and a potential technical approach in precision medicine.

In-vivo proton range verification for reducing the risk of permanent alopecia in medulloblastoma treatment

ObjectiveThe purpose of this work is the clinical commissioning of a recently developed in-vivo range verification system for the head treatment of pediatric medulloblastoma patients. Inaccurate beam range for such treatment could lead to either inadequate dose coverage of the target volume or excessive dose to the head skin resulting permanent alopecia. MethodsThe in-vivo range verification system is designed to perform pre-treatment range verification and adjustment. An array of Si-diode detectors is to be placed on the patient immobilization mask in the exit direction of a whole-brain field; signal is analyzed, and the extracted water equivalent path length (WEPL) is compared to the expected one, revealing if a range correction is needed. The method was tested in solid water and anthropomorphic head phantom, with validation based on independent WEPL measurements. The measured WEPL were compared to those computed by the treatment planning system (TPS). ResultsThe accuracy for the WEPL measurements by the diode system in both solid water and anthropomorphic head phantom were on average within a millimeter from more accurate measurement by the dose-extinction technique, with the error range for the two phantoms as (0–1 ​mm) and (0–1.3 ​mm), respectively. When compared to the WEPL calculated by the treatment planning system, the measured values were on average within 1 ​% (range 0–3%) of the beam range. The accuracy of dose measurements by the diodes in the fall-off part of the depth dose profile was validated against the reference Markus chamber. No need for further correction (due to different beam parameters and detector dose ageing effects) was found. ConclusionsThe range verification workflow was successfully tested in the anthropomorphic head phantom. The performance of the in-vivo range verification system and related workflow meet the clinical requirements in terms of the needed WEPL accuracy for pretreatment range verification.