Brain Tumors in Children Amar Gajjar Gregory H. Reaman Judy M. Racadio Franklin O. Smith Editors 123 Brain Tumors in Children Amar Gajjar • Gregory H. Reaman Judy M. Racadio • Franklin O. Smith Editors Brain Tumors in Children Editors Amar Gajjar Gregory H. Reaman Department of Oncology Children’s National Medical Center St. Jude Children’s Research Hospital George Washington University Memphis, TN Silver Spring, WA USA USA Judy M. Racadio Franklin O. Smith Cincinnati, OH Medpace USA Cincinnati, OH USA ISBN 978-3-319-43203-8 ISBN 978-3-319-43205-2 (eBook) https://doi.org/10.1007/978-3-319-43205-2 Library of Congress Control Number: 2018946696 © Springer International Publishing AG, part of Springer Nature 2018 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Printed on acid-free paper This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Preface Brain tumor programs are like orchestras—several components all working together seamlessly to provide optimal clinical care and conduct basic, trans- lational, and clinical research, and thus advance the field. The last decade has witnessed unprecedented advances in the field of neuro-oncology that have impacted the entire practice of treating children with brain tumors. Using modern molecular technologies that have facilitated a unique insight into the genomic make up of pediatric brain tumors, we have gained in-depth knowl- edge into the genetic heterogeneity of these tumors. This knowledge has also generated a new classification that is gradually being implemented by the World Health Organization (WHO). The fields of neurosurgery, neuroimag- ing, and radiation oncology have witnessed technological advances that have revolutionized how these modalities have been deployed in the treatment of children. The introduction of targeted therapies based on tumor molecular profiling has injected a new era of hope for curing brain tumors that were incurable in the past. Neurocognitive deficits, which are a significant concern in children treated for brain tumors, are being addressed with interventions that promise to remediate some of the damage. Long-term follow-up of brain tumor survivors has documented the unique health risk profile that these chil- dren carry for their life based on their treatment history. The recognition that more than two-thirds of the burden of pediatric cancer occurs in developing countries raises unique challenges regarding delivery of adequate therapy to this disadvantaged population. The authors of the individual chapters, all experts in their own domains, have done an outstanding job of succinctly documenting the recent advances and providing a glimpse of where the field is headed over the next few years. This book is a must-read for trainees, junior and seasoned practitioners in the field as it provides a lucid update in a rapidly emerging field. Memphis, TN, USA Amar Gajjar Silver Spring, WA, USA Gregory H. Reaman Cincinnati, OH, USA Judy M. Racadio Cincinnati, OH, USA Franklin O. Smith v Contents 1 Epidemiology of Pediatric Central Nervous System Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Nicholas A. Vitanza, Cynthia J. Campen, and Paul G. Fisher 2 Principles of Pediatric Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . 17 P. Ryan Lingo, Asim F. Choudhri, and Paul Klimo Jr 3 Principles of Radiation Oncology . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Shannon M. MacDonald, Ranjit S. Bindra, Roshan Sethi, and Matthew Ladra 4 Imaging Children with CNS Tumors . . . . . . . . . . . . . . . . . . . . . . . 65 Julie H. Harreld 5 Predisposition Syndromes to Central Nervous System Cancers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Alberto Broniscer and Kim Nichols 6 Modern Principles of CNS Tumor Classification . . . . . . . . . . . . . 117 Stefan M. Pfister, David Capper, and David T. W. Jones 7 Medulloblastoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Nicholas G. Gottardo and Christopher I. Howell 8 Ependymoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Hendrik Witt and Kristian W. Pajtler 9 High-Grade Glioma, Including Diffuse Intrinsic Pontine Glioma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 Matthias A. Karajannis, Matija Snuderl, Brian K. Yeh, Michael F. Walsh, Rajan Jain, Nikhil A. Sahasrabudhe, and Jeffrey H. Wisoff 10 Low-Grade Gliomas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Anna K. Paulsson, Michael A. Garcia, David A. Solomon, and Daphne A. Haas-Kogan 11 Germ Cell Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 Kee Kiat Yeo and Girish Dhall 12 Childhood Craniopharyngioma . . . . . . . . . . . . . . . . . . . . . . . . . . . 265 Thomas E. Merchant vii viii Contents 13 R are Embryonal Brain Tumours . . . . . . . . . . . . . . . . . . . . . . . . . 289 Adriana Fonseca, Salma Al-Karmi, Alexandre Vasiljevic, Andrew Dodghsun, Patrick Sin Chan, Lucie Lafay Cousin, Jordan Hansford, and Annie Huang 14 C ognitive Late Effects and Their Management . . . . . . . . . . . . . . 317 Heather M. Conklin, Jane E. Schreiber, and Ashley S. Fournier-Goodnight 15 L ong-Term Outcomes Among Survivors of Childhood Central Nervous System Malignancies: Late Mortality, Subsequent Neoplasms, Endocrine and Neurologic Morbidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347 Gregory T. Armstrong, Raja B. Khan, and Wassim Chemaitilly 16 I ntegrating Palliative Care into the Ongoing Care of Children with CNS Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379 Justin N. Baker 17 G lobal Challenges in Pediatric Neuro-Oncology . . . . . . . . . . . . . 403 Simon Bailey, Jeannette Parkes, and Alan Davidson 1 Epidemiology of Pediatric Central Nervous System Tumors Nicholas A. Vitanza, Cynthia J. Campen, and Paul G. Fisher 1.1 Introduction the molecular profile of pediatric CNS tumors has made it clear that epidemiology, viewed Tumors of the central nervous system (CNS) through a prism of genetics and epigenetics, can comprise a broad and diverse collection of neo- offer even greater insights into this incredibly plasms within pediatric oncology. Yet when taken challenging group of tumors. Epidemiology together pediatric brain and spine tumors repre- today considers not only environmental, parental, sent the most common childhood cancer with an and birth factors that may increase the risk of incidence of 5.57 per 100,000 annually and are a pediatric CNS tumors, but also germline and leading cause of cancer-related death in patients molecular features that are causal or pathogno- under 19 years of age (Ostrom et al. 2014; Siegel monic of tumor types and subtypes. et al. 2015). Factors such as genetic predisposi- tion, age, and sex play an increasingly significant role in understanding presentation, management, 1.2 Astrocytomas and Other and etiology of childhood brain tumors. Although Gliomas long-standing observations regarding general patterns of CNS tumors continue to be clinically The gliomas are a heterogeneous group of useful, the introduction of molecular subtypes, tumors, comprised mostly of astrocytomas. such as in medulloblastoma and ependymoma, Pediatric astrocytomas are divided into four and the discovery of epigenetic regulators, such grades by the World Health Organization (WHO), as in diffuse intrinsic pontine gliomas (DIPG) with pilocytic astrocytomas (WHO grade I) being and other diffuse midline gliomas with H3K27M the most common subtype of pediatric CNS mutations, have repurposed epidemiological tumor, comprising approximately 15% (Ostrom findings and reconceptualized CNS tumor clas- et al. 2014; Louis et al. 2007). The incidence of sification (Louis et al. 2016). The elucidation of pilocytic astrocytomas in children in England and the USA is 0.75–0.97 per 100,000, and these tumors have an exceedingly low incidence of N. A. Vitanza Division of Pediatric Hematology/Oncology, Seattle metastasis or malignant transformation (Ostrom Children’s Hospital, University of Washington School et al. 2014; Stokland et al. 2010; Fisher et al. of Medicine, Seattle, WA, USA 2008; Arora et al. 2009). Although they may C. J. Campen · P. G. Fisher (*) occur in any CNS location including the spine, Division of Child Neurology, Department of they most commonly arise from the posterior Neurology, Lucile Packard Children’s Hospital at fossa, optic pathway and hypothalamus, or brain Stanford, Stanford University, Palo Alto, CA, USA e-mail: [email protected] stem (Fernandez et al. 2003; Gajjar et al. 1997; © Springer International Publishing AG, part of Springer Nature 2018 1 A. Gajjar et al. (eds.), Brain Tumors in Children, https://doi.org/10.1007/978-3-319-43205-2_1 2 N. A. Vitanza et al. Hayostek et al. 1993; Khatib et al. 1994). Diffuse drome in NF1 patients, as well as the practice of astrocytomas (WHO grade II), anaplastic astro- asymptomatic surveillance imaging in that group, cytomas (WHO grade III), and glioblastomas while 90% of sporadic cases present with new (WHO grade IV) have an incidence of 0.27, 0.08, neurologic symptoms. Subependymal giant cell and 0.15 per 100,000 children 0–14 years of age, astrocytomas (SEGAs) are another WHO grade I respectively. Low-grade gliomas, which are com- astrocytoma subtype that develop almost exclu- prised of WHO grade I and II astrocytomas as sively in patients with tuberous sclerosis (TS), well as WHO grade I gangliogliomas, most com- which occurs in 1 in 5600 live births (O’Callaghan monly present with greater than 6 months of et al. 1998). Five to twenty percent of patients symptoms (Fisher et al. 2008). The incorporation with TS develop SEGAs, often in adolescence, of molecular characteristics in the 2016 WHO but congenital cases have also been reported classification of tumors of the CNS will assist in (Adriaensen et al. 2009; O’Callaghan et al. 2008; a deeper epidemiological understanding by Hahn et al. 1991). addressing distinct biologic entities, such as dif- Several WHO grade II subtypes can be distin- fuse gliomas with IDH mutations and diffuse guished by histology and presentation. midline gliomas with H3K27M mutations (Louis Pilomyxoid astrocytomas (WHO grade II) have a et al. 2016). more aggressive course than pilocytic astrocyto- Children with pilocytic astrocytomas have mas (WHO grade I), a greater propensity for excellent outcomes of >96% overall survival growing in the hypothalamochiasmatic region, (OS) at 10 years, and patients with subtotal resec- and often present earlier with a mean age of 3.3 tions do not do significantly worse than patients years (Bhargava et al. 2013). Pleomorphic xan- with gross total resections (Ostrom et al. 2014; thoastrocytomas (WHO grade II) are typically Gajjar et al. 1997). Posterior fossa tumors are located in the superficial temporal lobe; they common in children, with pilocytic astrocytomas classically present with seizures and have a being the second most common tumor arising in median age at diagnosis of 20.5 years and an that location, behind only medulloblastoma; approximately 75% overall survival (Gallo et al. mean age of occurrence is 7.1 years (Smoots 2013; Perkins et al. 2012). These can rarely trans- et al. 1998). Up to 60% of pilocytic astrocytomas form into a high-grade glioma. are associated with a KIAA1549:BRAF fusion, Low-grade gliomas of the brain stem can be which is associated with a better outcome (Becker pilocytic astrocytomas or gangliogliomas, which et al. 2015; Jones et al. 2008). Optic pathway and typically occur dorsally and have the possibility hypothalamic astrocytomas are most often pilo- of long-term cure. WHO grade II, III, and IV cytic astrocytomas, but other subtypes of low- gliomas of the brain stem have dismal outcomes grade gliomas also account for a small number of and together comprise diffuse intrinsic pontine cases (Hoffman et al. 1993; Laithier et al. 2003). glioma (DIPG). The 2016 WHO classification Optic pathway gliomas (OPGs) occur in approxi- has adjusted that nomenclature in favor of diffuse mately 15% of patients with neurofibromatosis midline gliomas, as diffuse gliomas of the pons, type 1 (NF1), though they most often occur spo- thalamus, and spinal cord may form a more bio- radically (Listernick et  al. 1989). OPGs are logically distinct category when H3K27M muta- reported to have a broad median age between 4.3 tions are present (Louis et al. 2016; Shankar et al. and 8.8 years, and those occurring in patients 2016). with NF1 present at a significantly earlier age DIPGs arise most commonly in the ventral than sporadic cases (Listernick et  al. 1989; pons and comprise 10–15% of all pediatric CNS Nicolin et al. 2009; Singhal et al. 2002; Ahn et al. tumors and 80% of brain stem gliomas, affecting 2006; Janss et al. 1995; Khafaga et al. 2003; roughly 300 children in the USA each year Jahraus and Tarbell 2006; Avery et al. 2011). The (Ostrom et al. 2014; Ramos et al. 2013; Smith variation in age of presentation may be secondary et  al. 1998). Males and females are affected to the presence of a cancer predisposition syn- equally and the median age of presentation is 1 Epidemiology of Pediatric Central Nervous System Tumors 3 7 years (Lassiter et al. 1971; Lober et al. 2014; et al. 2007). They account for 15% of CNS Veldhuijzen van Zanten et al. 2014). Presentation tumors in patients 0–14 years of age and 12% in usually consists of a classic triad of ataxia, cra- those 0–19 years of age, with incidences of 0.78 nial nerve palsies, and pyramidal tract signs and 0.64 per 100,000, respectively; these inci- developing over 1 month, although atypical cases dences have remained unchanged since at least can present more slowly over several months 1990 (Ostrom et al. 2014; Johnston et al. 2014). (Fisher et al. 2000). It is now recognized that Embryonal CNS tumors rarely occur outside of approximately 17% of patients undergo both childhood with the median age at presentation local and distant neuraxis dissemination by 15 being 7.3 years, and 44% of them being diag- months, which is not far beyond the median over- nosed between the ages of 4 and 9 years (Ostrom all survival of patients with DIPG, as only 10% of et al. 2014; Kool et al. 2012). Medulloblastomas, patients survive beyond 2 years and only 2–3% the most common malignant brain tumor in pedi- are considered long-term survivors (Gururangan atrics, histologically appear as PNETs specifi- et al. 2006; Hargrave et al. 2006; Jackson et al. cally arising in the posterior fossa (Northcott 2013). Recently, 80% of DIPGs have been found et  al. 2011). A minority of medulloblastoma to harbor mutations in K27M of histone 3.1 or cases have been reported in patients with genetic 3.3, which are associated with mutations in predisposition syndromes such as Gorlin, Turcot ACVR1 and p53, respectively (Taylor et al. 2014; B, Li–Fraumeni, ataxia telangiectasia, Nijmegen Wu et al. 2012). breakage, Rubenstein–Taybi, and Coffin–Siris High-grade gliomas (HGGs) occur much syndromes (Distel et al. 2003; Hart et al. 1987; more frequently in adults, with an increasing Larsen et al. 2014; Skomorowski et al. 2012; incidence with age to a peak between the ages of Taylor et al. 2001; Rogers et al. 1988). Overall, 75 and 85 years (Ostrom et al. 2014). The out- there is a male predominance of 1.5:1, with comes of patients with high-grade gliomas appear females reported to have superior outcomes, to be inverse to patient age, as 5-year overall sur- although again this is likely subgroup dependent, vivals for children less than three and those 3–14 as there are fewer females in the higher risk years of age are 31–66% and 19%, respectively Group 3 and 4, while more young females have (Mathew et al. 2014). Glioblastoma has been sonic hedgehog (SHH) driven tumors (Louis reported in classic CNS tumor predisposition et al. 2007; Northcott et al. 2011). Historically, syndromes, such as neurofibromatosis, Li– patients clinically classified as average-risk had a Fraumeni, and Turcot syndromes, as well as in 5-year OS of roughly 85%, while high-risk several genitourinary syndromes, such as Turner patients suffered poorer outcomes with near 70% and Mayer–Rokitansky–Küster–Hauser syn- OS and patients with large-cell anaplastic histol- drome, though the majority of cases are believed ogy had particularly dismal outcomes (Kool et al. to be sporadic (Hanaei et al. 2015; Jeong and Yee 2012; Gajjar et al. 2006; Packer et al. 2006; 2014; Macy et al. 2012; Gonzalez and Prayson Tarbell et al. 2013; Ramaswamy et al. 2013). 2013). Overall, long-term survival in patients with medulloblastoma is achieved in only 66% of patients, with 10% suffering from secondary 1.3 Embryonal Tumors malignancies, 32% of which are secondary brain tumors (Ning et al. 2015). Embryonal brain tumors are a diverse group of Although particular subsets of medulloblastoma aggressive neoplasms, including medulloblas- have long been suspected to behave differently, it is toma, primary neuroectodermal tumors (PNET), now commonly accepted that there are four distinct atypical rhabdoid/teratoid tumors (ATRT), and molecular subgroups: WNT, SHH, Group 3, and pineoblastoma, which share high mitotic activity Group 4, which account for 11%, 28%, 27%, and and a predilection for dissemination throughout 34% of cases, respectively (Kool et al. 2012; the neuraxis, and are all WHO grade IV (Louis Northcott et  al. 2011; Badiali et  al. 1991).