Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

An international clinical program for the diagnosis and treatment of children, adolescents and young adults with ependymoma

Final Version 3.1_ April 22 nd , 2020

Sponsor: Centre Léon Bérard Sponsor Protocol Number: ET-13-002 EudraCT Number: 2013-002766-39 VHP number: VHP358 (VHP201385)

Page | 1 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

SIGNATURE PAGE

SIOP EPENDYMOMA II – An international Clinical Program for the diagnosis and treatment of children, adolescents and young adults with ependymoma Protocol version 3.1 dated April 22 nd , 2020

This protocol has been approved by:

Name: Dr David PEROL Trial Role: Medical and scientific coordinator

Signature: ______Date: 22-Apr-2020

This protocol describes the SIOP Ependymoma Program II and provides information about procedures for patients taking part in this trial. The protocol should not be used as a guide for treatment of patients not taking part in the SIOP Ependymoma Program II.

CONFIDENTIALITY NOTE: This trial protocol is the property of the authoring Coordinating Investigators and is protected by copyright. The content of the trial protocol and the case report forms is confidential and oral or written disclosure to any uninvolved / third parties (without previous written agreement by one of the international or national coordinating investigators) is prohibited.

Page | 2 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

AMENDMENTS

The following amendments and/or administrative changes have been made to this protocol since the implementation of the first approved version.

Protocol Amendment Date of Type of version Summary of main modifications number amendment amendment number 1 August 21st 2.0 Substantial ° Inclusion and exclusion criteria 2014 amendment ° Objectives ° Guidelines regarding dose modification of IMPs ° Guidelines regarding dose of IMPs to be administered (stratum 3) ° Volume of IMP administration (Methotrexate and Carboplatin) ° Hydration guidelines regarding Cisplatin administration ° Radiotherapy guidelines ° Supportive treatments & Concomitant medications ° MRI time point ° Assessment of Response to Treatment ° Hearing – PTA Dose Modification ° Endocrine evaluation ° Biological material needed for BIOMECA studies ° Planned interim analysis and stopping rules (Stratum 2) ° Adverse Events reporting ° Withdrawal from trial treatment ° Withdrawal of consent to data collection ° Valproate PK/PD studies 2 3.0 Substantial ° Update of the study rationale amendment ° Study design (staging mandatory for all patients) ° Rearrangement of selection criteria ° Update of the WHO classification ° Inclusion of relapsing patients ° Addition of examinations (BARS, audiograms) ° Valproic acid administration schedule ° Update of the pharmacovigilance section ° Update of the Integrated BIOMECA study ° Precisions on timing of study examinations 3 3.1 Non- ° Change of Principal Investigator Coordinator substantial ° Recruitment period extended amendment ° Addition of Greece ° Contact updates

Page | 3 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

FLOW SHEETS OF TREATMENTS

Pre -Operative MRI + Surgery

FIRST STEP Staging Phase Study entry Post-Operative MRI + CSF

Central review of imaging and pathology Confirm diagnosis

Evaluate need of second look surgery

Patients ≥ 12 months Patients ≥ 12 months Patients < 12 months Patients not included in with no measurable residue with measurable and patients not eligible to one of the interventional inoperable residue receive RT studies Randomized phase III trial to evaluate the efficacy of post Randomized frontline Randomized phase II Observational study radiation maintenance phase II chemotherapy chemotherapy study: chemotherapy study and exploration of Alternated myelosuppressive (VEC CDDP for 16 weeks) the efficacy of a boost of and non myelosuppressive radiotherapy chemotherapy +/- HDACi (Valproate)

Stratum 1 Stratum 2 Stratum 3

SECOND STEP Interventional or observational Phase

Page | 4 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

FLOW SHEETS OF PATIENTS INCLUSION

MRI

Surgery

Entry in staging phase

1st Consent

Central review of imaging and pathology: Confirm diagnosis and evaluate second look surgery

2nd Consent Residual / operable 2nd Consent

Stratum 1 Absence of residual disease Second look surgery Inoperable residue Stratum 2 < 12 months or Age ≥ 12 months not eligible for radiotherapy Age ≥ 12 months and eligible for radiotherapy and e ligible for radiotherapy

2nd Consent

Stratum 3

Page | 5 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

STRATUM 1 Patients with no measurable residual disease (R0-1-2), WHO Grade II-III ependymoma No metastasis Age > 12 months and < 22 years Adequate bone marrow, renal and liver function.

Randomisation

Conformal RT 59.4 Gy (children < 18 months or with risk factors (*): 54 Gy) Daily fraction 1.8 Gy, 5 fractions/week

Observation Maintenance CT (**) WEEK 1 => WEEK 6 => WEEK 11=> WEEK 16 D1: Vincristine (VCR) 1.5 mg/m² D1-D3 : Etoposide (VP16) 100 mg/m² D1: Cyclophosphamide 3000 mg/m²

WEEK 4 => WEEK 9=> WEEK 14 D1: Cisplatin (CDDP) 80 mg/m² D1: Vincristine (VCR) 1.5 mg/m²

(*) multiple surgeries (more than 2) or poor neurological status. (**) dose adaptation for children less than 10 kg

Page | 6 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

STRATUM 2 Patients with measurable inoperable residual disease (R3-R4), WHO Grade II-III ependymoma No metastasis Age > 12 months and < 22 years Adequate bone marrow, renal and liver function

Randomisation

VEC -HD -MTX (**) VEC (**) WEEK 1=> WEEK 4=> WEEK 7 WEEK 1=> WEEK 4=> WEEK 7 VEC-MTX VEC D1: Vincristine (VCR) 1.5 mg/m² D1: Vincristine (VCR) 1.5 mg/m² D1-D3: Etoposide (VP16) 100 mg/m² D1-D3: Etoposide (VP16) 100 mg/m² D1: Cyclophosphamide: 3000 mg/m² D1: Cyclophosphamide 3000 mg/m²

WEEK 3=> WEEK 6=> WEEK 9 D1: Methotrexate at 8000 mg/m 2 as a 24 hour infusion. MRI with central r eview 2nd look surgery if possible

No residual Residual disease disease

Conformal RT 59.4 Gy (children < 18 months or with risk factors (*): 54 Gy) Daily fraction of 1.8 Gy, 5 fractions/week

Boost of RT of 8 Gy to residue (Daily fraction of 4 Gy: 2 fractions)

Maintenance CT (if no prior progression under VEC) (**) WEEK 1 => WEEK 6 => WEEK 11 => WEEK 16 D1: Vincristine (VCR) 1.5 mg/m² D1-D3 : Etoposide (VP16) 100 mg/m² D1: Cyclophosphamide: 3000 mg/m²

WEEK 4 => WEEK 9 => WEEK 14 D1: Cisplatin (CDDP) 80 mg/m² D1:Vincristine (VCR) 1.5 mg/m²

(*) multiple surgeries (more than 2) or poor neurological status. (**) dose adaptation for children less than 10 kg

Page | 7 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

STRATUM 3 Children < 12 months or those not eligible to receive radiotherapy

Adequate bone marrow, renal and liver function and ammonia

Randomisation

STANDARD CHEMOTHERAPY STANDARD CHEMOTHERAPY + HDACi = valproate

Maintenance HDACi Treatment for one year period If no progression during frontline chemotherapy

CHEMO +/- HDACi (**)

CYCLE N° 1 2 3 4 5 6 7

Vincristine - Carboplatin D1 D 57 D113 D169 D225 D281 D337

Vincristine - Methotrexate D15 D71 D127 D183 D239 D295 D351 Vincristine - Cyclophosphamide D29 D85 D141 D197 D253 D309 D365

Cisplatin 2-day D43 D99 D154 D211 D267 D323 D379 Continuous infusion 44 100 155 212 268 324 380 +/- Valproate (*) Initial dose: 30 mg /kg/day for two weeks in 2 divided doses (15mg/Kg/b.d.). Increasing weekly up to 40->50->60 mg /kg/day in 2 divided doses until serum target level achieved. If therapeutic levels of 100-150 µg/ml are not achieved when giving 60 mg /kg/day administer in three divided doses (20mg/kg/dose b.d.). If therapeutic level is still not achieved increase to higher total daily dose in 5 mg/kg/day increments that is 65 mg /kg/day then 70 mg /kg/day (23.3mg/kg/dose t.d.s) in 3 divided doses.

Dosing schedule (***) Dose for infants Dose for infants Dose for infants over 12 months 6 to 12 months less than 6 months

Vincristine 1.5 mg/m² x 1 1.125 mg/m² x 1 0.75 mg/m² x 1 (Maximum dose: 2mg) Carboplatin 550 mg/m²x 1 412.5 mg/m² x 1 275 mg/m² x 1 Methotrexate 8000 mg/m² x 1 6000 mg/m² x 1 4000 mg/ m² x 1 Cyclophosphamide 1500 mg/m² x 1 1125 mg/m² x 1 750 mg/m² x 1 Cisplatin 40 mg/m² x 2 30 mg/m² x 2 20 mg/m² x 2 Valproate * (BID) 30 mg/kg/day* 30 mg/kg/day* 30 mg/kg/day*

* Initial dosing then according to monitoring **If residual disease please consider for further surgery at each reassessment point. *** For patients aged: • 12 months and over: full surface-area-based dose of chemotherapy • 6-11 months: 75% of the surface-area-based dose of chemotherapy • Under 6 months: 50% of the surface-area –based dose of chemotherapy

Page | 8 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

TRIAL COORDINATION

OVERALL SIOP EPENDYMOMA II PROGRAM

Coordinating Dr Pierre LEBLOND investigator Institut d’Hématologie et d’Oncologie Pédiatrique 1 place Joseph Renaut 69673 Lyon Cedex, France Phone: +33 4 69 16 65 71 Fax: +33 4 78 78 27 09 Email: [email protected] Stratum 1 Phase III study for patients with no measurable residual disease ≥ 12 months of age Chief investigator / Dr Pierre LEBLOND Cécile DALBAN Chief statistician Institut d’Hématologie et d’Oncologie Direction de la Recherche Clinique et de Pédiatrique l’Innovation 1 place Joseph Renaut Centre Léon Bérard 69673 Lyon Cedex, France 28, rue Laënnec Phone: +33 4 69 16 65 71 69373 Lyon Cedex 08, France Fax: +33 4 78 78 27 09 Phone: +33 4 78 78 75 71 Email: [email protected] Fax: +33 4 78 78 27 15 Email : [email protected] Stratum 2 Phase II study for patients with measurable inoperable residual disease ≥ 12 months of age Chief investigator / Dr Maura MASSIMINO Dr Rosalba MICELI Chief statistician Pediatric Unit Fondazione IRCCS Istituto Nazionale dei Fondazione IRCCS Istituto Nazionale dei Tumori Tumori Via Venezian 1 20133 Milano, Italy Via Venezian 1 20133 Milano, Italy Phone: +39 02-23903198 Phone: +39 0223902593 Fax: +39 02-23902095 Fax: +39 0223902648 email: [email protected] email: [email protected] Responsible for the Dr Martin ENGLISH randomised Birmingham Children’s Hospital chemotherapy part Steelhouse Lane Birmingham B4 6NH, United Kingdom Phone: +44 121 333 8412 Fax: +44 121. 333 8241 Email: [email protected] Responsible for the Dr Lorenza GANDOLA radiotherapy part Fondazione IRCCS Istituto Nazionale dei Tumori (boost) Via Venezian 1 20133 Milano, Italy Phone: +39 0223902471 Email: [email protected] Stratum 3 Primary chemotherapy strategy for children < 12 months of age or ineligible to receive radiotherapy

Page | 9 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Chief investigator / Professor Richard GRUNDY Pr Keith WHEATLEY Chief statistician Children's Brain Tumour Research Centre D 32 Cancer Research UK Clinical Trials Unit Medical School (CRCTU) Queen's Medical Centre University of Birmingham Nottingham Edgbaston NG7 2UH, United Kingdom Birmingham Phone: +44 (0) 115 8230620 B15 2TT, United Kingdom Fax: +44 (0) 115 8230696 Phone: +44 (0)121 415 9119 Email: [email protected] Fax: +44 (0)121 414 3700 Email: [email protected]

Page | 10 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

NATIONAL COORDINATION

Austria Dr Martin Benesch Division of Pediatric Hematology/Oncology Department of Pediatrics and Adolescent Medicine Medical University of Graz Auenbruggerplatz 38 8036 Graz, Austria Phone: +43-316-385-80427 - Fax: +43-316-385-13450 Email: [email protected]

Belgium Dr Caroline Piette Département Hémato-oncologie pédiatrique CHR de la Citadelle Boulevard du 12ème de Ligne, 1, 4000 Liège, Belgium Phone: +32 4 225 60 97 - Fax: +32 4 225 76 87 Email: [email protected]

Czech Republic Dr Jaroslav Sterba Pediatric Oncology Department University Hospital Brno Cernopolni street 9, PC 613 00, Czech Republic Phone: +420 532 234 600/755 - Fax: +420 532 234 614 Email: [email protected]

Denmark Dr Pernille Wendtland Edslev Pediatric oncology Aarhus University Hospital Brendstrupgaardsvej 100, 8200 Aarhus, Denmark Phone: +45 78451701 - Fax: +45 78451710 Email: [email protected]

Finland Dr K ir sti Sirkiä Division of Hematology - Oncology and Stem Cell Transplantation Children's hospital Helsinki University Central Hospital Haartmaninkatu 4, PO Box 372, FIN 00029 Hus, Finland Phone: +358 (0)40 6464323 Email: [email protected]

France Dr Pierre LEBLOND Institut d’Hématologie et d’Oncologie Pédiatrique 1 place Joseph Renaut, 69673 Lyon Cedex 08, France Phone: +33 4 78 69 16 65 71 - Fax: +33 4 78 78 27 09 Email: [email protected]

Germany Dr Stefan Rutkowski Department of Pediatric Haematology and Oncology University Medical Center Hamburg-Eppendorf, Martinistrabe. 52, 20246 Hamburg, Germany Phone: +49 040-741054270 - Fax: +49 040-741059063 Email: [email protected]

Page | 11 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Greece Dr Vita Ridola Department of Pediatric Hematology and Oncology Children's Hospital MITERA 6 Erithrou Stavrou , 15123 Athens, Greece Phone: +30 210 6869632 Email: [email protected] Ireland Dr Michael Capra Our Lady’s Children’s Hospital Haematology and Oncology department Crumlin, Dublin 12, Ireland Phone : +353 1 409 6659 - Fax : +351 1 456 3041 Email: [email protected]

Italy Dr Maura Massimino Fondazione IRCCS Istituto Nazionale dei Tumori 20133 Milan, Italy Phone: +39 0223902593 - Fax: +39 0223902648 Mail: [email protected]

The N etherland s Dr Jasper van der Lugt Princess Maxima Center for pediatric oncology P.O. Box 85090 .3508 EA Utrecht Heidelberglaan 25 .3584 CS Utrecht, The Netherlands Phone: +31 6 1855 96 94 - Fax: +31 88 97 294 52 Email: [email protected]

Norway Dr Ingrid Kristin Torsvik Haukeland University Hospital Department of Pediatrics Jonas Lies Vei 79.5053 Bergen, Norway Phone: +47 5597 5199/+47 5597 3859 - Fax : +47 5597 1509 Email: [email protected]

Poland Dr Marta Perek -Polnik Klinika Onkologii IP-CZD The Children's Memorial Health Institute CMHI (IP-CZD) Al. Dzieci Polskich 20, 04 -730 Warsaw, Poland Phone: +48 22 815 17 79 - Fax: +48 22 815 75 75 Email: [email protected]

Slovenia Dr Lidija Kitanovski University Medical Center Ljubljana Dpt. of Paediatrics - Div. of Haematooncology Bohoriceva 20, 1000 Ljubljana, Slovenia Phone: + 386 1 522 9215 / 522 9256 - Fax: + 386 1 522 4036 Email: [email protected]

Spain Dr. Ana Fernández -Teijeiro Pediatríc Oncology Program Hospitales Universitarios Virgen Macarena y Virgen del Rocío Avda. Dr Fedriani n*3, 41071 Sevilla, Spain Phone : +34677903132 - Fax : +34954370892

Page | 12 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Email: [email protected]

Sweden Dr Helena Mörse Pediatric Oncology Dept, Skane University Hospital 221 85 Lund, Sweden Phone: +46 46 178281 - Fax: +46 46 130573 Email: [email protected]

Page | 13 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Switzerland Dr Nicolas Gerber Department of Oncology University Children’s Hospital Steinwiesstrasse 75, CH-8032 Zurich, Switzerland Phone: +41 44 266 31 17 - Fax: +41 44 266 34 61 Email: [email protected]

United Kingdom Dr Richard Grundy Children's Brain Tumour Research Centre D 32 Medical School, Queen's Medical Centre, Nottingham, NG7 2UH, United Kingdom Phone: +44 115 8230620 - Fax: +44 115 8230696 Email: [email protected]

The national trial coordinators will have the oversight for the conduct of the study within their national group and will also sit on the Study’s Steering Group that will oversee the running of the study as a whole.

Each National Coordinating Centre will appoint a national radiotherapy coordinator or designated person who will have the oversee for Radiotherapy issues within their National Coordinating Centre, and in particular will be responsible for the organisation of Radiotherapy Quality Control procedures.

Each National Coordinating Centre will appoint a national reference centre for biology and pathology. The respective Coordinator or designated person will have the responsibility to undertake, coordinate the quality control of the molecular diagnostics, central pathology review, and translational biological studies.

Each National Coordinating Centre will appoint a national reference centre for neuroradiology . The respective Coordinator or designated person will have the responsibility to undertake the coordination and the quality control of the central neuroradiological review.

Each National Coordinating Centre will appoint a national reference centre for neurosurgery . The respective Coordinator or designated person will have the responsibility to conduct, coordinate central neuroradiological review and advice for surgery outcome evaluation and second look surgery.

Each National Coordinating Centre will appoint a national QoS coordinator or designated person who will have the oversight for QoS issues within their National Coordinating Centre, and in particular will be responsible for the organisation of QoS assessment procedures.

Each National Coordinating Centre will appoint a national neuropsychological coordinator or designated person who will have the oversight for neuropsychological issues within their National Group, and in particular will be responsible for the organization of neuropsychological assessment procedures and data collection.

Page | 14 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

TRIAL SYNOPSIS TITLE SIOP Ependymoma Program II: An International Clinical Program for the diagnosis and treatment of children, adolescents and young adults with ependymoma

SPONSOR Centre Léon Bérard

EUDRACT 2013-002766-39 NUMBER VERSION 3.1 dated April 22 th , 2020 INVESTIGATOR - Dr Pierre LEBLOND COORDINATOR STUDY DESIGN The Ependymoma Program is a comprehensive program to improve the accuracy of the primary diagnosis of ependymoma and explore different therapeutic strategies in children, adolescents and young adults, accordingly. This program is opened to all patients diagnosed with ependymoma below the age of 22 years. It will include a centralised review of pre and post-operative imaging to assess the completeness of the resection. It will also include a central review of pathology to confirm the histological diagnosis. The biological markers 1q gain, Tenascin C status, RELA-fusion, YAP fusion, H3.3K27me3 and molecular subgroup by methylation array will be prospectively assessed for prospective evaluation of disease subgroups. Further biological evaluations will be coordinated within the integrated BIOMECA study. After surgery and central review of imaging and pathology, patients will be offered the opportunity to undergo second look surgery, if possible. Patients will be enrolled in one of 3 different strata according to the outcome of the initial surgical resection (residual disease vs no residual disease), their age or eligibility / suitability to receive radiotherapy. These 3 different strata correspond to 3 therapeutic strategies according to the patient status. 1. Stratum 1 is designed as a randomised phase III study for patients who have had a complete resection, with no measurable residual disease (as confirmed by centrally reviewed MRI) and are ≥ 12 months and < 22 years at diagnosis. Those patients will be randomised to receive conformal radiotherapy followed by either 16 weeks of chemotherapy with VEC+CDDP, or observation.

2. Stratum 2 is designed as a randomised phase II study for patients who have inoperable measurable residual disease and who are ≥ 12 months and < 22 years at diagnosis. Those patients will be randomised to two different treatment schedules of chemotherapy either with VEC or VEC+ high dose methotrexate (VEC +HD-MTX). After completion of the frontline chemotherapy, patients will be assessed for response (MRI) and will receive second look surgery when feasible. For those patients who remain unresectable with residual disease despite frontline chemotherapy and for whom second line surgery is not feasible, there will be a study of the safety of a radiotherapy boost of 8 Gy that will be administered to the residual tumour immediately after the completion of the conformal radiotherapy. Patients without evidence of residual disease after the chemotherapy and/or a second look surgery are not eligible for radiotherapy boost . All patients who have

Page | 15 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

not shown progression under chemotherapy will receive, as maintenance therapy, a 16 week course of VEC+CDDP following completion of radiotherapy. 3. Stratum 3 is designed as a randomised phase II chemotherapy study in children <12 months of age or those not eligible to receive radiotherapy. These patients will be randomised to receive a dose dense chemotherapy alternating myelosuppressive and relatively non-myelosuppressive drugs at 2 weekly intervals, with or without, the addition of the histone deacetylase inhibitor, valproate. Observational study: After staging phase, patients that do not fulfil the inclusion criteria of one of the interventional strata will be enrolled and followed up via an observational study which will be analysed descriptively.

PURPOSE The overall aim of this project is to improve the outcome of patients diagnosed with ependymoma by improving and harmonising the staging and the standard of care of this patient population and to improve our understanding of the underlying biology thereby informing future treatment. The program will evaluate new strategies for diagnosis (centralized reviews of pathology and imaging) and new therapeutic strategies in order to develop treatment recommendations. Patients will be stratified into different treatment subgroups according to their age, the tumour location and the outcome of the initial surgery. Each subgroup will be studied in a specific randomised study to evaluate the proposed therapeutic strategies.

Stratum 1: In patients that have undergone complete resection and radiotherapy, the 5 years Progression Free Survival (PFS) is only 60%. Survival following relapse after primary treatment is generally very poor and there is thus a need to improve tumour control in order to postpone disease relapse. Based on the available data, it is still debated whether maintenance chemotherapy might improve or not the tumour control in patients ≥ 12 months of age at diagnosis after complete resection and standard radiotherapy. The aim of the stratum 1 is to evaluate the clinical impact of 16-week chemotherapy regimen with VEC+CDDP following surgical resection and conformal radiotherapy in terms of progression free survival in patients with completely removed intra cranial ependymoma. The proposed chemotherapy schedule is a combination of VEC and CDDP, drugs that have currently the best response rate.

Stratum 2: This arm is designed as a phase II trial for patients with residual disease to investigate the possible activity of HD-MTX by giving to all patients the benefit of VEC chemotherapy whilst randomising half of patients to receive additional HD-MTX. This will generate clinical data to decide whether methotrexate should be investigated in future phase III trials. A negative result will be as useful as a positive result as it would support decisions to reduce the use of HD-MTX in the infant population and will support initiatives to try alternative therapies. All patients will receive cRT. For patients who remain with a residual inoperable disease after induction chemotherapy and cRT, an 8 Gy boost of radiotherapy to the residual tumour will be delivered immediately after the end of the cRT. This boost of radiotherapy will be administered with the hope of maintaining local control and extending overall survival whilst also maintaining the patient’s quality of life. Safety of boost radiotherapy will be also evaluated.

Page | 16 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Stratum 3 This stratum is designed as a phase II trial to evaluate the benefit of postoperative dose intense chemotherapy administered alone or in combination with valproate. The aim is to minimize the risk of drug resistance whilst maximizing the intensity of treatment in very young children. The information collected about the outcome of infants enrolled in this study will be of considerable benefit and to informing future international studies for the ependymoma population.

Integrated Biological Study The SIOP Ependymoma Program II supports the identification of informative prognostic biomarkers within the collaborative BIOMECA study. This high priority initiative is an essential element of the overall program to improve future treatment of ependymoma OBJECTIVES Overall program: The primary objective : to determine whether the assessment of residual disease can be improved by a centralized review of post-operative MRI and whether such review increases the rate of complete resection compared to historical controls. Does central neurosurgical and radiological review increase resection rates?

Secondary objectives: • To evaluate second look surgery rates as compared to historical controls.

Exploratory objective: • To test the hypothesis that key molecular events are predictive of clinical behaviour of ependymoma. • To prospectively evaluate 1q copy-number status, Tenascin C, RELA-fusion, YAP fusion, H3.3K27me3 and molecular subgroup (by methylation array) as prognostic and predictive biomarkers in ependymoma within clinical trial setting in addition to a number of exploratory markers. This part of the program will be organised by the European Ependymoma Biology Consortium called “Biomarkers of Ependymomas in Children and Adolescents (BIOMECA)” throughout a cooperation agreement with the aim to identify informative prognostic biomarkers for the assessment of disease status and predictive response to therapy: • To identify new biomarkers for clinical and biological behaviour of ependymoma (location, recurrence, chemo resistance, invasion, metastasis). • To validate known biomarkers in the context of the new international prospective study for newly diagnosed ependymoma. • To provide a new comprehensive grading scheme based on the histological, and immunohistochemical and/or biological criteria that could be used for newly diagnosed ependymoma. • To validate and compare the techniques used to assess biomarkers for further stratification of patients • To select the best biomarker(s) and establish a prognostic signature for ependymoma.

Stratum 1: Patient with no measurable residual disease and ≥12 months of age (phase III) The primary objective is to test the hypothesis that there will be an improvement in progression-free survival in patients who receive 16 weeks chemotherapy (VEC+CDDP) following surgical resection and conformal radiotherapy when compared to those that

Page | 17 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

undergo surgical resection and radiotherapy alone.

Secondary objectives: • To evaluate whether overall survival (OS) is improved in patients who receive 16 weeks chemotherapy (VEC+CDDP) following surgical resection and radiotherapy when compared to those that undergo surgical resection and radiotherapy alone. • To compare the neuroendocrine morbidity in each treatment arm. • To evaluate the neuropsychological morbidity in each treatment arm. • To evaluate the quality of survival (QoS) in each treatment arm. • To determine the safety and the tolerance to 16 weeks chemotherapy (VEC+CDDP) following surgical resection and radiotherapy when compared to those that undergo surgical resection and radiotherapy alone.

Stratum 2: Patients with inoperable measurable residual disease and ≥12 months of age (phase II) The primary objective : To compare the activity of 2 post-operative chemotherapy schedules, VEC or VEC+HD-MTX in patients who have incompletely resected tumour.

Secondary objectives: • To determine the safety and tolerability to chemotherapy VEC + HD-MTX following surgical resection when compared to those that receive VEC alone • To evaluate whether OS is improved in patients who receive VEC + HD-MTX following surgical resection when compared to those that receive VEC alone. • To evaluate whether progression-free survival (PFS) is improved in patients who receive VEC + HD-MTX following surgical resection when compared to those that receive VEC alone. • To compare the neuroendocrine morbidity in each treatment arm. • To evaluate the neuropsychological morbidity in each treatment arm. • To evaluate the quality of survival in each treatment arm. • To determine safety of 8 Gy boost radiotherapy in patients with residual disease despite frontline chemotherapy and after 59.4 Gy conformal radiotherapy.

Stratum 3: Randomized phase II chemotherapy study in children < 12 months of age or those not eligible to receive radiotherapy Primary objective : To evaluate the progression free survival in children unable to receive radiation therapy and who receive valproate, as a histone deacetylase inhibitor in addition to the primary chemotherapy strategy when compared to those that undergo chemotherapy without valproate.

Secondary objectives: • To evaluate whether OS is improved in patients by adding valproate to primary chemotherapy strategy when compared to those that receive primary chemotherapy only. • To evaluate whether radiotherapy-free survival is improved in patients by adding valproate to primary chemotherapy strategy when compared to those that receive primary chemotherapy only. • To compare the neuroendocrine morbidity in each treatment arm, • To evaluate the neuropsychological morbidity in each treatment arm. • To evaluate the QoS in each treatment arm. • To determine the safety and tolerability of valproate added to the primary chemotherapy strategy for children unable to receive radiation therapy.

Page | 18 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

ENDPOINTS Overall program: Primary endpoint measure: Gross Total Resection (GTR) rate Secondary endpoints measure: Second look surgery rate

Only descriptive statistics will be produced for GTR rate and second look surgery rate.

Stratum 1: Primary endpoint measure: PFS from the date of randomisation to the date of event defined as progression or death due to any cause. Secondary endpoints measures: OS measured from the date of randomisation to the date of death due to any cause. QoS Neuropsychological outcomes Neuroendocrine outcomes (Neuroendocrine late effects) Short and long term safety: Adverse Events (CTCAE v4.03)

Stratum 2: Primary endpoint measure: Number of treatment responders. Objective response to chemotherapy is measured based on SIOP-E Neuro Imaging guidelines. Secondary endpoint measures: OS measured from the date of randomisation to the date of death due to any cause. PFS from the date of randomisation to the date of event defined as progression or death due to any cause. QoS Neuropsychological outcomes Neuroendocrine outcomes (Neuroendocrine late effects) Short and long term safety: Adverse Events (CTCAE v4.03) Exploratory endpoints measures: Toxicity monitoring in the subgroup receiving radiotherapy boost. Event-free survival in patients receiving a radiotherapy boost.

Stratum 3: Primary endpoint measure: Progression-free survival from the date of randomisation to the date of event defined as progression or death due to any cause. Secondary endpoint measures: OS measured from the date of randomisation to the date of death due to any cause. Radiotherapy-free survival rate QoS Neuropsychological outcomes Neuroendocrine outcomes (Neuroendocrine late effects) Short and long term safety : Adverse Events (CTCAE v4.03) Exploratory Endpoints measures (optional): • Pharmacokinetic modelling will be carried out using Valproate pharmacokinetic parameters in conjunction with patient characteristics and clinical parameters in order to investigate the key factors involved in determining individual Valproate drug exposures within the patient population. Valpro ate pharmacodynamics will be followed t hroughout , monitoring changes in

Page | 19 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

histone H3 and H4 acetylation. Changes between baseline and time of steady state valpro ate will be correlated by trough levels and clinical response. SAMPLE SIZE A minimum of 480 patients will be enrolled in the overall program according the following statistical considerations: Stratum 1- Patients with no measurable residual disease and ≥ 12 months of age (phase III) This stratum is designed to detect an improvement of the 5-year PFS rate of 15% within the chemotherapy arm as compared to the observation arm (75% versus 60%). Assuming 10 years of recruitment, 13 years for total follow-up (i.e 3 years for the last included patient), and two interim futility analyses, 160 patients per arm (109 events in total) are required with a power of 85% and a bilateral alpha risk of 5%. • a total of 320 eligible patients will be randomised • randomisation 1:1 stratified by tumour location (supra versus infratentorial), grading (classic versus anaplastic) and age (≥ or < 3 years old at diagnosis) into the two treatment arms (i.e. 160 patients per arm) • Expected accrual: 10 years

Stratum 2 – Patients with inoperable measurable residue and ≥ 12 months of age (phase II) This stratum has been designed as a single stage randomised phase II Jung design. Due to the rarity of the target population, the sample size is limited. It is anticipated that during the 10 years of recruitment, it will be possible to randomise at least 60 patients - 30 patients per arm. With 30 patients per arm randomised, under the assumption that the true response rates are 30% in the control arm and 45% in the methotrexate arm then the suggested analysis rule below will give 79% power to successfully carry methotrexate forward for further evaluation, but with a 34% risk that it will be carried forward when in fact it is ineffective. • A total of 60 eligible patients will be randomised • Randomisation 1:1 into the two treatment arms (i.e. 30 patients per arm) • Expected accrual: 10 years

Stratum 3 – Randomized phase II chemotherapy study in children <12 months of age or those not eligible to receive radiotherapy This stratum aims to recruit a minimum of 100 patients over 10 years, with 2.5 years follow up at the end of the study. Median follow-up will therefore be 5 years. Assuming that the true 5-year PFS rates are 45% and 60% in the control and valproate arms, respectively, then the study will have 80% power to detect the difference between the arms at a 1-tailed alpha level of 0.25. If recruitment is less than target, and only 82 patients are recruited, then the study retains 75% power to detect this difference. Conversely, if the study manages to recruit 130 patients, then it will have 80% power to detect the difference at a 1-tailed alpha level of 0.2. • A minimum of 100 eligible patients will be randomised • Randomisation 1:1 into the two treatment arms (i.e. 50 patients per arm) • Expected accrual: 10 years

STUDY Stratum 1: Patients with no measurable residual disease and ≥ 12 months TREATMENTS Conformal Radiotherapy (c RT): Any patient ≥ 12 months will receive cRT of 59.4 Gy (only 54 Gy if < 18 months or with risk factors such as multiple surgeries (more than 2) or poor neurological status ) with a 0.5 cm margin in a 1.8 Gy daily fraction s, 5 fractions per week

Page | 20 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

(under general anesthesia, if required), followed by either: 16-weeks of chemotherapy VEC+CDDP within 6 weeks after completion of standard 59.4 Gy radiotherapy or observation.

Stratum 2: Patients with inoperable measurable residue and ≥ 12 months Postoperative induction chemotherapy: Patient ≥12 months with a residual measurable disease after surgery will receive an 8-week course of chemotherapy: VEC+HD-MTX or VEC. Chemotherapy should be started within 3 weeks of surgery (any delay should be discussed with both the chief investigator coordinator in the country concerned and the chief investigator coordinator of the program). Second look surgery: Patients will be evaluated by MRI for second look surgery whenever possible, Conformal Radiotherapy (c RT): All patients will receive cRT of 59.4 Gy (only 54 Gy if less than 18 months old or with risk factors as multiple surgeries (more than 2) or poor neurological status) with a 0.5 cm margin in a 1.8 Gy daily fraction, 5 fractions per week (under general anesthesia if required). Boost of conformal radiotherapy (if residual tumour): Patients with tumours that persist despite pre-radiation chemotherapy and conformal radiotherapy will receive a boost of 8 Gy of cRT to the residual tumour bed (2 fractions of 4 Gy on 2 consecutive treatment days). 16-week maintenance Chemotherapy: All patients will receive 16-week maintenance chemotherapy with VEC+CDDP provided they showed no prior progression under first chemotherapy.

Stratum 3: Randomized phase II chemotherapy study in children<12 months or those not eligible to receive radiotherapy: Patients will be randomized to receive 2 different schedules of chemotherapy. The complete course of chemotherapy comprises 4 administrations of alternating myelosuppressive and relatively non-myelosuppressive drugs repeated every 56 days for a total of 7 cycles alone or in association with valproate. At the end of the chemotherapy, patients who have received valproate will receive valproate for an additional year. Chemotherapy should be started as soon as recovered from surgery and ide ally within 3 weeks of surgery. MAIN Inclusion criteria for overall program (staging phase): INCLUSION • Main residence in one of the participating countries, CRITERIA • Age < 22 years old at diagnosis, • Histological diagnosis of intracranial or spinal, localized or metastatic, ependymoma according to local pathologist (all WHO grades) including: myxopapillary ependymoma, ependymoma (papillary, clear-cell, tanycytic), ependymoma RELA- fusion positive or anaplastic, • Delivery to national referral pathology center of formalin-fixed paraffin embedded (FFPE) tumour tissue blocks (or at least twenty 5 µm sections on charged slides with sufficient interpretable material and at least ten 10 µm curls in an Eppendorf tube) • Written informed consent (staging) for collection and transfer of biological samples. • All patients and/or their parents or legal guardians willing and able to comply with protocol schedule and agree to sign a written informed consent. • Patients must be affiliated to a Social Security System in countries where this is mandatory.

Page | 21 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Inclusion and exclusion criteria have been defined for each stratum of the program (See section 4. Eligibility ) After staging phase, patients that do not fulfil the inclusion criteria of one of the interventional strata will be enrolled and followed up into an observational study and descriptive analysis will be performed. EXPECTED Enrolment period: 10 years STUDY PERIODS Treatment duration: • stratum 1: 7 months • stratum 2: 9 months • stratum 3: 2 years and 2 months Follow up: Patients of each stratum will be followed 5 years after treatment completion. Whole study duration : • stratum 1: 15 years and 7 months • stratum 2: 15 years and 9 months • stratum 3: 17 years and 2 months Main analysis will be performed 3 years (stratum 1), 0.5 year (stratum 2) and 2.5 years (stratum 3) after the final patient inclusion. Long term evaluation for all patients: PFS and OS evaluation will be performed at a minimum of 5 years after end of therapy. Beyond the completion of the 5 year follow up, all investigations will be performed according to local practice and are not required by protocol. However, the data generated from these additional examinations could be collected.

Based on the results observed, an updating of OS and PFS data may be requested by the steering committee beyond study duration. Long term evaluation for children under 18 years old at the end of follow up: ° A neuropsychological assessment and Quality of survival evaluation will be performed at 18 years of age for children below 18 at the completion of the 5 year follow up. ° PFS and OS evaluation will be performed at until 18th year of age. Consequently, if the last patient enrolled is under 1 year of age, study data could be collected over a maximum time period of 28 years. STUDY SITES This will be a multicentre international program which will involve the following countries: Austria, Belgium, Czech Republic, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Norway, Poland, Slovenia, Spain, Sweden, Switzerland and United Kingdom .

Page | 22 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

ABBREVIATIONS

AE ADVERSE EVENT ALT ALANINE AMINOTRANSFERASE ANC ABSOLUTE NEUTROPHIL COUNT ANSM AGENCE NATIONAL DE SECURITE DU MEDICAMENT - COMPETENT AUTHORITY FOR FRANCE AR ADVERSE REACTION AST ASPARTATE AMINOTRANSFERASE AUC AREA UNDER THE CURVE BAER BRAINSTEM AUDITORY EVOKED POTENTIALS BEV BEAMS EYE-VIEW BID BIS IN DIE (TWICE DAILY) BIOMECA BIOMARKERS OF EPENDYMOMA IN CHILDREN AND ADOLESCENTS BM BONE MARROW BP BLOOD PRESSURE CA COMPETENT AUTHORITY CBCL CHILD BEHAVIOR CHECK LIST CDDP CISPLATIN CI CHIEF INVESTIGATOR CNS CENTRAL NERVOUS SYSTEM COG CHILDREN’S ONCOLOGY GROUP CPM CYCLOPHOSPHAMIDE CPT CONTINUOUS PERFORMANCE TEST CR COMPLETE RESPONSE CRCTU CANCER RESEARCH UK CLINICAL TRIALS UNIT (UNIVERSITY OF BIRMINGHAM) CRF CASE REPORT FORM cRT CONFORMAL RADIOTHERAPY CR UK CANCER RESEARCH UK CT COMPUTED TOMOGRAPHY CTC COMMON TERMINOLOGY CRITERIA CTCAE COMMON TERMINOLOGY CRITERIA FOR ADVERSE EVENTS CTV CLINICAL TARGET VOLUME DAMPA 4-[[(2,4-DIAMINO-6-PTERIDINYL) METHYL] METHYLAMINO]-BENZOIC ACID DCRI DIRECTION DE LA RECHERCHE CLINIQUE ET DE L’INNOVATION DICOM DIGITAL IMAGING AND COMMUNICATION IN MEDICINE DLT DOSE LIMITING TOXICITY DRCI DIRECTION DE LA RECHERCHE CLINIQUE ET DE L’INNOVATION DRR DIGITALLY RECONSTRUCTED RADIOGRAPHY DSUR DEVELOPMENT SAFETY UPDATE REPORT DTPA DIETHYLENE TRIAMINE PENTAACETIC ACID DVH DOSE VOLUME HISTOGRAMS EOT END OF TREATMENT EU EUROPEAN UNION FFPE FORMALIN-FIXED PARAFFIN EMBEDDED GCP GOOD CLINICAL PRACTICE G-CSF GRANULOCYTE COLONY STIMULATING FACTOR

Page | 23 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

GFR GLOMERULAR FILTRATION RATE GP GENERAL PRACTICTIONER GTR GROS TUMOUR RESECTION GTV GROSS TUMOUR VOLUME HDAC HISTONE DEACETYLASE HDACi HISTONE DEACETYLASE INHIBITOR HD-MTX HIGH DOSE METHOTREXATE hTERT HUMAN TELOMERASE REVERSE TRANSCRIPTASE IB INVESTIGATOR BROCHURE iDSMC INDEPENDENT DATA SADETY MONITORING COMMITTEE IMP INVESTIGATIONAL MEDICINAL PRODUCT IMRT INTENSITY MODULATED RADIATION THERAPY ISF INVESTIGATOR SITE FILE IV INTRAVENOUS MFD MAXIMUM FEASIBLE DOSE MHRA MEDICINES AND HEALTHCARE PRODUCTS REGULATORY AGENCY MRI MAGNETIC RESONANCE IMAGING MTD MAXIMUM TOLERATED DOSE MTX METHOTREXATE NICR NORTHERN INSTITUTE FOR CANCER RESEARCH OAR ORGAN AT RISK OS OVERALL SURVIVAL PD PHARMACODYNAMIC PFS PROGRESSION FREE SURVIVAL PHGG PAEDIATRIC HIGH GRADE GLIOMA PI PRINCIPAL INVESTIGATOR PK PHARMACOKINETIC PRBC PACKED RED BLOOD CELLS PTV PLANNING TARGET VOLUME QC QUALITY CONTROL QoL QUALITY OF LIFE QoS QUALITY OF SURVIVAL RBC RED BLOOD CELLS RNA RIBONUCLEIC ACID RT RADIOTHERAPY SAE SERIOUS ADVERSE EVENT SIOP INTERNATIONAL SOCIETY OF PAEDIATRIC ONCOLOGY SPC SUMMARY OF PRODUCT CHARACTERISTICS SUSAR SUSPECTED UNEXPECTED SERIOUSADVERSE REACTION TMG TRIAL MANAGEMENT GROUP TSC TRIAL STEERING COMMITTEE VEC VINCRISTINE – ETOPOSIDE - CYCLOPHOSPHAMIDE VCR VINCRISTINE VPA VALPROATE VP16 ETOPOSIDE WBC WHITE BLOOD CELL

Page | 24 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

TABLE OF CONTENTS SIGNATURE PAGE ...... 2 AMENDMENTS ...... 3 FLOW SHEETS OF TREATMENTS ...... 4 FLOW SHEETS OF PATIENTS INCLUSION ...... 5 TRIAL SYNOPSIS ...... 15 ABBREVIATIONS ...... 23 TABLE OF CONTENTS ...... 25 1 BACKGROUND AND RATIONALE...... 30 1.1 Background ...... 30 1.2 Trial Rationales ...... 32 1.2.1 Current clinical results ...... 32 1.2.2 Rationale for a European program ...... 39 1.2.3 Rationale to evaluate key molecular events ...... 39 1.2.4 Rationale for randomisation of a maintenance VEC+CDDP chemotherapy ...... 40 1.2.5 Rationale for randomizing high dose methotrexate ...... 41 1.2.6 Rationale for a boost of radiotherapy ...... 43 1.2.7 Rationale for randomizing valproate in young children receiving front line chemotherapy ...... 43 1.2.8 Rationale to evaluate neuropsychological and neurocognitive outcomes...... 46 1.2.9 Rationale for registering metastatic and relapsing patients ...... 47 1.2.10 Rationale for registering spinal ependymomas ...... 47 1.2.11 Benefit and Risk assessments ...... 47 2 AIMS, OBJECTIVES AND OUTCOME MEASURES ...... 50 2.1 Aims and objectives ...... 50 2.1.1 Overall program ...... 50 2.1.2 Treatment stratum 1: Patients with no measurable residual disease and ≥ 12 months of age (phase III) . 51 2.1.3 Treatment stratum 2: Patients with inoperable measurable residue and ≥ 12 months of age (phase II) .. 51 2.1.4 Treatment stratum 3: Randomised phase II chemotherapy study in children < 12 months of age or those not eligible to receive radiotherapy (phase II) ...... 51 2.1.5 Observational study ...... 52 2.2 Outcome measures ...... 52 2.2.1 Overall program ...... 52 2.2.2 Treatment stratum 1: Patients with no measurable residual disease and ≥ 12 months of age (phase III) 52 2.2.3 Treatment stratum 2: Patients with inoperable measurable residue and ≥ 12 months of age (phase II) .. 52 2.2.4 Treatment stratum 3: Randomized phase II chemotherapy study in children <12 months of age or those not eligible to receive radiotherapy (phase II) ...... 53 3 TRIAL DESIGN ...... 54 3.1 Initial local procedure ...... 55 3.2 National Central Radiological Review ...... 55 3.3 National Central pathological review ...... 56 4 ELIGIBILITY ...... 57 4.1 Staging Phase ...... 57 4.1.1 Inclusion criteria ...... 57 4.1.2 Exclusion criteria ...... 57 4.2 STRATUM 1 ...... 57 4.2.1 Inclusion criteria ...... 57 4.2.2 Exclusion criteria ...... 58 4.3 STRATUM 2 ...... 59

Page | 25 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Inclusion criteria ...... 59 4.3.1 ...... 59 4.3.2 Exclusion criteria ...... 60 4.4 STRATUM 3 ...... 61 4.4.1 Inclusion criteria ...... 61 4.4.2 Exclusion criteria ...... 62 4.5 Observational study ...... 62 4.5.1 Inclusion criteria ...... 62 4.5.2 Exclusion criteria ...... 63 5 CONSENT ...... 63

5.1 SCREENING ...... 63 5.2 INFORMED CONSENT ...... 63 6 TRIAL ENTRY AND RANDOMISATION ...... 64

6.1 TRIAL PROCEDURES ...... 64 6.2 ONLINE RANDOMISATION PROCEDURES ...... 65 7 TREATMENT DETAILS ...... 66

7.1 STRATUM 1: PATIENTS WITH NO MEASURABLE RESIDUAL DISEASE AND ≥ 12 MONTHS OF AGE (P HASE III) ...... 66 7.1.1 Conformal radiotherapy ...... 66 7.1.1.1 Timing of radiotherapy ...... 67 7.1.1.2 Equipment ...... 67 7.1.1.3 Target volumes ...... 67 7.1.1.4 Dosimetry ...... 69 7.1.1.5 Treatment technique ...... 70 7.1.1.6 Dose to Organs at Risk ...... 70 7.1.1.7 Quality Control of Radiotherapy (QC) ...... 72 7.1.1.8 Definition of protocol deviation ...... 73 7.1.1.9 Treatment modifications due to hematological Toxicity ...... 73 7.1.2 Maintenance chemotherapy ...... 74 7.1.2.1 Vincristine ...... 75 7.1.2.2 Etoposide...... 75 7.1.2.3 Cyclophosphamide ...... 75 7.1.2.4 Cisplatin ...... 76 7.1.2.5 Investigations during chemotherapy ...... 76 7.2 STRATUM 2: PATIENTS WITH INOPERABLE MEASURABLE RESIDUE AND ≥ 12 MONTHS OF AGE (P HASE II) ...... 77 7.2.1 Definition or Investigational Medicinal Products (IMPs) ...... 77 7.2.2 VEC+ High dose Methotrexate arm ...... 77 7.2.3 VEC arm ...... 78 7.2.3.1 Recommended guidelines for the administration of drugs ...... 79 7.2.3.1.1 Vincristine ...... 79 7.2.3.1.2 Etoposide ...... 79 7.2.3.1.3 Cyclophosphamide ...... 79 7.2.3.1.4 Methotrexate ...... 80 7.2.3.2 Investigations during chemotherapy ...... 83 7.2.4 Conformal radiotherapy ...... 83 7.2.5 Radiotherapy boost ...... 83 7.2.5.1 Requirements to deliver the boost ...... 84 7.2.5.2 Quality Control of Radiotherapy (QC) ...... 85 7.2.5.3 Definition of protocol deviation ...... 85 7.2.5.4 Treatment Modifications due to hematological Toxicity ...... 85 7.2.6 Maintenance chemotherapy ...... 86 7.2.6.1 Vincristine ...... 86 7.2.6.2 Etoposide...... 87 7.2.6.3 Cyclophosphamide ...... 87 7.2.6.4 Cisplatin ...... 87

Page | 26 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.2.6.5 Investigations during chemotherapy ...... 88 7.3 STRATUM 3: RANDOMIZED PHASE II CHEMOTHERAPY STUDY IN CHILDREN < 12 MONTHS OR THOSE NOT ELIGIBLE TO RECEIVE RADIOTHERAPY ...... 89 7.3.1 Definition or Investigational Medicinal Products (IMPs) ...... 89 7.3.2 Post-operative intensive Chemotherapy ...... 89 7.3.2.1 Vincristine ...... 91 7.3.2.2 Carboplatin ...... 91 7.3.2.3 Methotrexate ...... 91 7.3.2.4 Cyclophosphamide ...... 93 7.3.2.5 Cisplatin (Note this is different to the regime in VEC-Cisplatin) ...... 93 7.3.2.6 Valproate ...... 94 7.3.2.7 Investigations during chemotherapy ...... 97 7.3.2.8 Indication for radiotherapy ...... 97 8 MODIFICATION OF DOSE CHEMOTHERAPIES ...... 98 8.1 Hematology ...... 98 8.2 Neurotoxicity of vincristine ...... 98 8.3 Nephrotoxicity of Cisplatin: ...... 99 8.3.1 Cisplatin dose modification used with VEC in Strata 1 and 2 ...... 99 8.3.2 Cisplatin dose modification used in Stratum 3 ...... 100 8.4 Ototoxicity of platinum-based chemotherapies ...... 100 8.5 Toxicity of methotrexate ...... 101 9 SUPPORTIVE TREATMENTS & CONCOMITANT MEDICATIONS ...... 101 10 ASSESSMENTS ...... 102

10.1 MEDICAL HISTORY , PHYSICAL AND NEUROLOGICAL EXAMINATIONS ...... 102 10.2 LOCAL AND CENTRAL PATHOLOGICAL ASSESSMENTS ...... 102 10.3 BIOLOGICAL ASSESSMENTS ...... 103 10.3.1 Hematology ...... 103 10.3.2 Blood chemistry ...... 103 10.3.3 Urinalysis ...... 103 10.4 RADIOLOGICAL ASSESSMENTS ...... 103 10.4.1 Cranial MRI ...... 104 10.4.2 Spinal MRI ...... 104 10.4.3 Early postoperative imaging ...... 104 10.4.4 Definitions of residual tumour ...... 105 10.4.5 Assessment of Response to Treatment ...... 106 10.5 ENDOCRINE EVALUATION ...... 107 10.6 AUDIOGRAM OR BAER (B RAINSTEM AUDITORY EVOKED RESPONSES ) ...... 107 10.7 OPHTHALMOLOGY EVALUATION ...... 107 10.8 QUALITY OF SURVIVAL & QUALITY OF LIFE ASSESSMENT ...... 108 10.8.1 Quality of Survival: Questionnaires-studies ...... 108 10.8.2 Quality of Life Assessments (in a closer sense) ...... 108 10.9 NEUROPSYCHOLOGICAL AND NEUROCOGNITIVE ASSESSMENTS ...... 110 10.10 INTEGRATED BIOLOGICAL STUDY BIOMECA...... 113 10.10.1 Samples to be collected ...... 114 10.10.2 Objectives of the BIOMECA study ...... 114 10.11 VALPROATE PK/PD PROFILE ...... 117 10.12 PATIENT FOLLOW -UP DURING TREATMENT ...... 117 10.13 PATIENT FOLLOW UP AFTER TREATMENT (ONLY FOR INTERVENTIONAL STRATA ) ...... 117 10.13.1 Short and medium term patient follow up after treatment ...... 117 10.13.2 Long-term patient follow up after treatment ...... 117 11 PATIENT WITHDRAWAL ...... 128

11.1 WITHDRAWAL FROM TRIAL TREATMENT ...... 128 11.2 WITHDRAWAL OF CONSENT TO DATA COLLECTION ...... 128

Page | 27 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

12 CLINICAL SAFETY DATA MANAGEMENT ...... 128

12.1 GENERALITIES ...... 128 12.1.1 Definitions (Directive 2001/20/EC of the European Parliament) ...... 128 12.1.2 Intensity criteria ...... 130 12.1.3 New safety issues ...... 130 12.2 REPORTING REQUIREMENTS ...... 130 12.2.1 Events that require immediate reporting on a Serious Adverse Event Form ...... 130 12.2.2 Events that do not require an immediate reporting on a SAE form ...... 130 12.2.3 Events that do not require SAE report ...... 131 12.2.4 Pregnancies ...... 131 12.2.5 Reporting periods ...... 132 12.3 REPORTING PROCEDURE ...... 132 12.3.1 Adverse Events ...... 132 12.3.2 Serious Adverse Events ...... 132 12.3.3 Reporting to the Competent Authority and main Research Ethics Committee ...... 134 13 DATA HANDLING AND RECORD KEEPING ...... 135

13.1 DATABASE AND DATA COLLECTION ...... 135 13.2 ARCHIVING ...... 135 14 QUALITY MANAGEMENT ...... 136

14.1 SITE SET -UP AND INITIATION ...... 136 14.2 MONITORING ...... 136 14.3 AUDIT AND INSPECTION ...... 137 14.4 NOTIFICATION OF SERIOUS BREACHES ...... 137 15 END OF TRIAL DEFINITION ...... 138 16 STATISTICAL CONSIDERATIONS ...... 138

16.1 OVERALL PROGRAM ...... 138 16.2 TREATMENT STRATUM 1: PATIENTS WITH NO MEASURABLE RESIDUAL DISEASE AND ≥ 12 MONTHS OF AGE (P HASE III) ...... 138 16.2.1 Sample size calculation ...... 138 16.2.2 Disease Stratification ...... 138 16.2.3 Analysis populations ...... 138 16.2.4 Statistical methods ...... 138 16.2.5 Planned interim analysis and stopping guidelines ...... 139 16.2.6 Planned final analysis ...... 139 16.3 TREATMENT STRATUM 2: PATIENTS WITH INOPERABLE MEASURABLE RESIDUE AND ≥ 12 MONTHS OF AGE (P HASE II) ...... 139 16.3.1 Sample size considerations for the randomised phase II treatment stratum ...... 139 16.3.2 Statistical methods for the randomised phase II treatment stratum ...... 140 16.3.3 Planned subgroup analysis...... 140 16.3.4 Planned interim analysis and stopping guidelines ...... 140 16.3.5 Planned main analysis...... 141 16.4 TREATMENT STRATUM 3: RANDOMISED PHASE II CHEMOTHERAPY STUDY IN CHILDREN < 12 MONTHS OF AGE OR THOSE NOT ELIGIBLE TO RECEIVE RADIOTHERAPY (P HASE II) ...... 142 16.4.1 Sample size considerations ...... 142 16.4.2 Statistical methods ...... 142 16.4.3 Planned subgroup analyses...... 142 16.4.4 Planned interim analysis and stopping guidelines ...... 142 16.4.5 Planned main analysis...... 142 17 TRIAL ORGANISATION AND STRUCTURE ...... 143

17.1 SPONSOR ...... 143 17.2 COORDINATING CENTRES ...... 143 17.3 RELATIONSHIP OF THE TRIAL COMMITTEES ...... 143 17.3.1 Trial Management Group ...... 143

Page | 28 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

17.3.2 Trial Steering Committee ...... 144 17.3.3 Independent Data Safety and Monitoring Committee (iDSMC) ...... 144 18 FINANCIAL CONSIDERATIONS ...... 145 19 ETHICAL CONSIDERATIONS ...... 145 20 CONFIDENTIALITY AND DATA PROTECTION ...... 146 21 INSURANCE ...... 147 22 PUBLICATION POLICY ...... 147 APPENDIX 1 – INTERNATIONAL CONTACT DETAILS ...... 148 APPENDIX 2 – WMA DECLARATION OF HELSINKI ...... 149 APPENDIX 3 – TOXICITY CRITERIA OF ADVERSE EVENTS V 4.03 (CTCAE)...... 150 APPENDIX 4 – SUMMARY OF PRODUCT CHARACTERISTICS ...... 151 APPENDIX 5 – BIOMARKERS OF EPENDYMOMA IN CHILDHOOD AND ADOLESCENCE (BIOMECA) ...... 152 APPENDIX 6 – VALPROATE PHARMACOKINETICS AND PHARMACODYNAMICS STUDIES (OPTIONAL) ...... 162 APPENDIX 7 – DEFINITION OF HIGHLY EFFECTIVE METHODS OF CONTRACEPTION ...... 164 REFERENCE LIST ...... 165

Page | 29 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

1 BACKGROUND AND RATIONALE

1.1 Background

Ependymomas are one of the most frequent malignant brain tumours in children and adolescents. Their prognosis, especially in young children remains poor and their effective treatment remains one of the most difficult tasks in paediatric oncology, particularly as half of all cases are diagnosed under 5 years of age [1]. The prognosis at relapse is ultimately dismal [2, 3]. Ependymomas are thought to arise from the lining of the ventricles and the central canal in the spine, though this is not universally the case. In childhood 90% are intracranial. They are usually associated with the ventricular cavities but can occur anywhere in the neuroaxis. Two thirds of tumours occur in the posterior fossa where they are intimately associated with the fourth ventricle. Dissemination by cerebrospinal fluid (CSF) is reported in 7 – 22% of cases [4-19]. Consistent histological grading of ependymomas is a difficult task and several classification systems have been proposed. The most frequently used system, World Health Organization (WHO) 2007[20], recognizes two principal variants in intracranial ependymoma of children: classic ependymoma (Grade II) and anaplastic ependymoma (Grade III). However, the distinction between classic and anaplastic is controversial. A clear international consensus on this difficult issue is clearly needed [21]. To date, there are no reliable biological prognostic markers for ependymoma. The role of histologic and molecular subtypes in prognostication remains controversial [21-26]. Further histological analysis combined with molecular biomarkers are needed to define further the histogenetic features of ependymoma to be used as prognostic markers and also as therapeutic targets in the hope of discovering novel therapeutic agents. The most common presentation is non-specific and related to raised intracranial pressure from obstructive hydrocephalus, but patients may also present with cerebellar or lower cranial nerve dysfunction. In young children, lethargy and irritability may be the only presenting symptoms. The outcome of intracranial ependymoma remains poor. After 5 years, progression free survival (PFS) rates range from 23 to 74%, while mortality is reported in up to 40% of affected children [27, 28]. This is particularly true for patients with incomplete resection. A number of prognostic factors have been identified in different studies. The degree of surgical resection is considered as the most consistent prognostic factor for children with ependymoma, but the post-surgical status needs to be confirmed by MRI scan within 24-48 hours post-surgery. In patients with no residual tumour, the 5-year PFS are 50 to 80% while those figures drop to 0 to 26% in patients with incompletely resected tumours [27, 28]. St Jude Children’s Research Hospital published the results of a US mono centre study [29] assessing a far more intensive surgical approach than most European studies. 56% of gross total resection (GTR) at first surgery was converted to 81% GTR after referral and entry into the St Jude trial protocol by offering re-operations to 43% of patients. Another European experience of a more intensive surgery approach was also recently published by an Italian group in the AIEOP study [14]. In this study, 110 patients were recruited. After initial surgery, 60% achieved GTR while 26% were re operated. After second look surgery, 14 patients achieved GTR raising thus the total resection rate from 60 to 73%. In this study, second look surgery proved feasible with no major morbidity and local tumour control was comparable to that of children with complete removal after the first operation. Today in Europe, GTR rates after initial surgery are around 50% and there is no general rule to offer patients second look surgery to attempt further resection prior to starting radiotherapy and / or chemotherapy.

Page | 30 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

However, the neurological costs of an aggressive surgery should however be considered [30] and surgical expert advice should be requested prior to undertaking second look surgery if needed. As higher morbidity is not unlikely with second, third or even more surgeries [30], there is a need to consider the level of experience of the neurosurgeon in this field. In some cases, referral of the patient to a more experienced centre to either advise or carry out the surgery may be most appropriate. Whether or not organising a central review of imaging and involving key opinion leaders in Europe (neurosurgeon and radiologist and chief investigator) to give recommendations on when it is appropriate to pursue more complete resection with further surgery with the aim of more patients achieving GTR, and thereby improving OS and PFS, remains a matter of debate. It is generally accepted that adjuvant therapy is required even when complete resection is achieved. Choices over adjuvant therapy are difficult, and to an extent, have depended on the underlying philosophies of national groups and institutions. Presently, there are marked differences in the post-surgical management of patients. This protocol seeks to determine a standardised European approach. Whereas focal radiotherapy to the tumour bed is the standard adjuvant treatment for ependymoma in older children and adults, immediate post-operative radiation therapy directed at the primary site in very young children (< 3 to 5 years) is not widely accepted due to the potential neurocognitive and neuroendocrine sequelae [9, 11, 31]. Historically, the standard dose of radiotherapy has been 54 Gy in 30 fractions to the planning target volume (PTV). More recent data from St Jude’s Children Research Hospital [32] has shown improved local control, event-free survival (EFS) and overall survival (OS) when 59.4 Gy in 33 fractions is given to the PTV using conformally planned radiotherapy. There was no apparent increase in neurocognitive late effects in this group of patients after 5 years of follow-up [33]. Consequently, this increased irradiation dose is now recommended. Since the early 1990s, chemotherapy has been widely used in the management of brain tumours in young children to defer radiotherapy and thus improve intellectual outcome. The perception of unacceptable side-effects of cranial radiotherapy in young children led a number of institutions to adopt chemotherapy based strategies to avoid or delay radiotherapy [9, 11, 15, 31, 34]. Over the last few years, there has been little improvement, if any, in the outcome of patients with ependymoma. In light of recent advances in neurosurgical techniques, radiotherapy and molecular diagnostics, there is an urgent need to incorporate and combine current practices to re-evaluate standard of care in a more uniform approach to provide a more favourable future for patients with ependymoma. Difficulties in predicting tumour behaviour from clinical and histological factors have shifted the focus to the molecular and cellular biology of ependymoma in order to identify new correlates of disease outcome and novel therapeutic targets [35] In this program we will investigate key molecular events in ependymoma tumorigenesis and test the hypothesis that improvements in the risk stratification using a combination of clinical and biological variables will enhance prediction of outcome. This study will also explore new predictive and prognostic biomarkers including a new proposal for histological grading recently proposed by a consensus panel of neuropathologists. The techniques to assess these markers will be prospectively compared and the most robust and valid methods determined for further use in the stratification of patients.

Page | 31 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

In this program, each national committee shall be composed at least of one reference radiologist, one reference neurosurgeon and the national coordinating investigator or designated cover. These committees will be responsible for confirming whether removal is complete, for giving advice on considerations for second look surgery and referral, if needed, and to help decide the stratum in which the patient should participate. Having such committees set up at European levels should allow improving GTR rate above 50%. As part of a potentially more intensive surgical strategy, there is also a need to monitor patients for increased risk of neurological morbidity in the context of a large well controlled clinical program. This protocol is also designed to cover diverse situations according to age, site and post-operative status. It will ask randomised questions in each situation. Particularly, the question of the role of post irradiation chemotherapy in patients with complete removal is addressed. The best induction chemotherapy in those with post-operative inoperable residue and the feasibility of a radiation boost will be evaluated. Finally, the role of HDAC inhibitors in infants receiving chemotherapy will be asked. Moreover, any patient that does not enter in one of these three randomised protocols will be included in an European observational study of ependymoma.

1.2 Trial Rationales

1.2.1 Current clinical results They are summarised in the following tables: for babies (i.e. < 3 years old) in Table 1 and older children in Table 2.

Page | 32 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

TABLE 1: Summary of literature for babies with Ependymoma (* extrapolated from curves):

PROTOCOLS TREATMENTS PFS PFS OS OS

PFS PFS OS atients atients Reference Reference Age range range Age (months) patients N° of eceiving eceiving (p) (p) RT Period 2 year 5 year 2 year 5 year Multivariate % p r Induction: 5 cycles A: VCR/CDDP/CY/VP 0-36 1993- residue Geyer CCG9921 74 38* 32 82* 72 >50 or B: VCR/CBDCA/IFO/VP 1997 (univ: ,020) [36] Maintenance: 8 cycles VCR / VP/ CBDCA/CY RD or PD → RT Induction: 5 cycles CDDP/ VCR/VP/ CY/±MTX Zacharoulis HD-CBDCA/TTT+VP and PBSC 1991- [37] Head Start 0,9-105 29 35* 12 70* 38 RT: if < 3y RD after treatment 2002 Venkatramani or > 3y + ST residue after treatment [38] or > 3y + PF VETOPEC Induction: < 48 1991- 14 29 36 <50 White [39] 4 cycles VCR /VP/ HD-CY 1995 Maintenance: 3 cycles VCR+CY/VP+CDDP/VCR+CY/VP+CBDCA

Page | 33 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

PROTOCOLS TREATMENTS Age range range Age (months) N° of patients Reference (p) (p) RT Period PFS 2 year PFS 5 year OS 2 year OS 5 year Multivariate % patients receiving SSK 87 Induction (for high risk only): PCB, Ifos, VP, MTX, CDDP, ARAC Maintenance: PCB+VCR, MTX+VCR U:

HIT87/92 RT: systematic when reach 3 year 1987- degree of Timmermann 1-33 34 35* 65* 62 Anaplastic HIT 92: 1997 resection [40] Induction: (0,07) CY+VCR+ITMTX,MTX+VCR+ITMTX,MTX+VCR+ITM TX,CBDCA+VP+ITMTX

PD → RT 7 cycles: PFS A: CBDCA, PCB 1990- (0,0004) BB SFOP B: VP, CDDP 5- 62 73 33 79 53 Grill [4] 1998 Residue C: VCR, C (0,0009) PD → RT

7 Cycles : (89) A: CBDCA, VCR data HR for B: MTX, VCR 1992- on death: dose CCLG 9204 1-93 47 41 79 63 56* Grundy [11] C: CY, VCR 2003 80 intensity D: CDDP non 0,04 PD → RT meta

Page | 34 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

PROTOCOLS TREATMENTS Age range range Age (months) N° of patients Reference (p) (p) RT Period PFS 2 year PFS 5 year OS 2 year OS 5 year Multivariate % patients receiving Merchant If residue personnal St Jude 7 weeks CBDCA, VP/VCR+Cy < 36 77 78* 72* 100 Communication RT focal systematic 2007 Induction: 4 cycles VCR+MTXHD +CY alternate CDDP + VP 1994- Massimino AIEOP 3-36 41 27 36 or VEC 2003 [15] RD or PD after treatment → RT

ARAC= aracytin, CBDCA= carboplatin, CDDP= cisplatin, CY = cyclophosphamide HD= high dose, ITMTX= intrathecal methotrexate, MTX= methotrexate, PBSC= peripheral blood stem cell rescue, PCB= procarbazine, TTT=Thiotepa, VCR= vincristine, VP=Etoposide, IFO= ifosfamide, RD =residual disease, PD =progression disease, RT: radiotherapy, R= randomisation.

Page | 35 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

TABLE 2: Summary of literature for children with Ependymoma (* extrapolated from curves)

PROTOCOLS TREATMENTS year PFS PFS year Age range range Age (years) N° of patients (p) (p) Period 2 PFS 5 year OS 2 year OS 5 year Multivariate Reference AIEOP #1 1993- 96 HFRT (70.4 Gy, [2 x 1,1Gy]) 3-21 46 78* 65 82 no residue 2001 * U: PFS Resid 0,05, Massimino grading 0,0008 [14] AIEOP #1 2 cycles VEC (VCR, VP, CY) >3 and 1993- 88 17 66* 35 61 residue HFRT (70.4 Gy [2x 1.1 Gy]) <21y 2001 *

AIEOP #2 2002- No residue RT 59,4Gy [1,8Gy] >1 48 84 98 2014 and grade II 5-year PFS=65% AIEOP #2 RT 59,4Gy [1,8Gy] (grading 0,012) No residue <21 62 62 79 4 cycles VEC (VCR, VP, CY) and grade III Massimino [97] 5-year OS=81% 1-4 cycles VEC (residue 0,09 and +/- second look surgery grading 0,09) AIEOP #2 + 59,4Gy 50 53 67 Residue + 8 Gy boost (2x4Gy)

Page | 36 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

PROTOCOLS TREATMENTS Age range range Age (years) N° of patients (p) (p) Period PFS 2 year PFS 5 year OS 2 year OS 5 year Multivariate Reference 3y 2 cycles IFO+ VP + MTX + CDDP + ARAC 59 HIT 88/89- CSI (35.2, boost 20 Gy) univ 1989- (83 if 85 3y* Timmermann 91 or 3 -16 71 70* meta 0,0001 1997 no * 75 [41] ANAPLAST. CSI (35.2, boost 20 Gy) resid 0,004 residue 8 cycles MTX, CDDP, VCR vs 38) RT varied: HFRT 70.7Gy [2 x 1Gy] Pilot or CF 54 Gy [1 x 1.8Gy] 1990- Univ Needle 19 85* 54 Needle +/- CSI 3-14 1992 PF 0,036 [42]

4 cycles: CBDCA+VCR Ifos+VP OS: grade (p=0·0052), Degree of resection (p<0·0001), (If residue 7 weeks CBDCA+VP/VCR + Cy) 1997- EFS 7 y EFS Ethnic (p=0·018). Merchant St Jude 0.9 - 22 153 94 7 y 81% 2007 85* 69% EFS grade [29] RT 59.4 Gy focal systematic * (p=0·008), Degree of resection (p<0·0001]), Gender (p=0·042)

Page | 37 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

PROTOCOLS TREATMENTS

Age range range Age (years) N° of patients (p) (p) Period PFS 2 year PFS 5 year OS 2 year OS 5 year Multivariate Reference

HFRT: 60 Gy [2x 1Gy] 1996 - 85 SFCE >5 24 75* 54 74 Conter [8] if complete removal, 66 Gy otherwise 2002 *

No difference between patients Garwin [43] 1995 - EFS CCG 9942 chemo +RT+ VCR, CY, CIS, VP 3-21 84 71% with residue 1999 57% (chemo/RT) vs no residue (RT)

ARAC= aracytin, CBDCA= carboplatin, CDDP= cisplatin, HD= high dose, ITMTX= intrathecal Methotrexate, MTX= methotrexate, PBSC= peripheral blood stem cell rescue, PCB= procarbazine, TTT=thiotepa, VCR= vincristine, VP=etoposide, IFO= ifosfamide, RD =residual disease, PD =progression disease, RT: radiotherapy, R= randomisation.

Page | 38 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

1.2.2 Rationale for a European program To date, there have been numerous publications about therapeutic interventions for ependymoma. However, because of inconsistencies of practice and infrequent trials, cohorts are usually too small to conclude on the risks and benefits of the intervention and results need to be confirmed in large randomised clinical trials to finally provide new recommendations. The standard treatment for intracranial ependymoma is surgery followed by radiation therapy in older children, and by chemotherapy in those who cannot receive radiation. The introduction of new medical practices can only be considered through large randomised clinical trials allowing reliable conclusions. Today, there is no trial open in Europe for patients with ependymoma. A nationwide protocol on ependymoma, built on a similar structure as the one proposed here, is currently on-going in the United States. In the following chapters, the rationale for innovative proposals will be outlined.

1.2.3 Rationale to evaluate key molecular events Difficulties in predicting tumour behaviour from clinical and histological factors [21] have shifted the focus to interrogating the molecular and cellular biology of ependymoma in order to identify new markers of clinical outcome and potentially novel therapeutic targets. Although our understanding of ependymoma biology has advanced, the development of novel prognostic classifications and targeted therapies is still required to enhance patient outcome for this tumour group, particularly in children. Although several candidates have been proposed, markers showing reproducible results in sizeable groups of young ependymoma patients are still required [44]. Indeed, several purported biologic prognostic markers in ependymoma have been shown to lose this capacity when assessed across clinical trial cohorts [45], highlighting the importance of analysing any prospective marker in standardised therapeutic groups. To date four biomarkers have been evaluated in retrospective cohorts of Childhood ependymoma in training and test sets and are sufficiently robust to include in this study as mandatory markers; a series of prospective discovery biomarkers will also be evaluated as part of the BIOMECA program (section 10.10). Mandatory markers include 1q gain [22], Tenascin-C (Anderiuolo F et al. in press) [46], RELA-fusion, YAP fusion, H3.3K27me3 and molecular subgroup by methlyation array as a global assessment. These can all be assessed using FFPE material. In this study, similarly to what is happening for similar tumours such as medulloblastomas, the best prospectively validated biological prognostic criteria, 1q gain, Tenascin C, NELL2& LAMA2, RELA- fusion, YAP fusion, H3.3K27me3 and molecular subgroup (by methylation array) will be utilised alongside clinical and histopathological indices and any newly identified marker from the BIOMECA study, in order to define the low-risk biological profile among ependymomas as well as to identify cases with high-risk biological profiles from the overall cohort. Prospective analysis of the large number of cases will help to:

1. Identify clinically relevant disease subgroups; characterize differential genomic signatures, and improving our understanding of the molecular basis of ependymoma,

2. correlate patients response to conventional treatments with specific molecular features in order to identify molecular markers that are valid surrogates of response to treatments; leading to the identification of independent and informative prognostic markers for a more efficient patient stratification in future SIOP clinical trials.

Page | 39 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

3. provide experimental evidence to guide the future use of chemotherapeutics as well as novel treatments targeting activated oncogenic pathways; the identification of targets for therapeutic exploitation. In addition to the mandatory biomarkers above BIOMECA has agreed an initial exploratory and discovery set of markers for validation within this prospective trial. BIOMECA provides organisational structure to test discovery biomarkers as well novel markers and enhanced understanding of this enigmatic disease. Advancing our understanding of the biology of ependymoma remains crucial to identify additional prognostic markers, discover molecular targets for novel or existing therapeutic agents in the clinic and allow adjuvant therapy to be tailored according to tumour specific molecular characteristics. Progress in these areas could minimize the long term adverse effects of therapy and improve patient survival A new histological grading scheme has been proposed following review of previous trial cohorts and now needs to be validated prospectively to assess histopathological features of the samples and explore the relevance of this new ependymoma grading scheme. It takes into account the cell density, the mitotic activity and the microvascular proliferation. Necrosis and atypia are no longer considered relevant for grading.

1.2.4 Rationale for randomisation of a maintenance VEC+CDDP chemotherapy The role for chemotherapy for intracranial ependymoma remains uncertain [47]. Timmermann et al. reported on a group of 28 children affected by anaplastic ependymoma who achieved complete resection and had 3-years EFS above 83% receiving radiotherapy followed by maintenance chemotherapy [41]: these results remain the best in terms of disease control in this subset of patients. A prospective study by Needle et al [48] used irradiation followed by carboplatin and vincristine alternating with ifosfamide and etoposide in patients greater than 36 months of age with newly diagnosed ependymoma. The 5-year PFS for the 10 patients with incompletely resected tumours was 80% and extent of surgical resection was not found to be of prognostic significance. Needle’s excellent survival statistics for incompletely resected ependymoma suggest that there may be a role for chemotherapy. Unfortunately, radiation therapy was not standardized and a portion of the patients received hyperfractionated radiation therapy, which confounds the analysis. An Italian protocol conducted by AIEOP adopted a VEC schedule containing etoposide, vincristine, and moderately high-doses of cyclophosphamide, aimed to improve dose intensity and thereby overcoming the chemoresistance of ependymoma [49] and obtaining a better local control in children with residual disease. Results documented a potential role of VEC in ependymoma with an objective response rate reaching 54% (95% confidence interval, 25%–81%) [14]. Chemotherapy consisted of vincristine (1.5 mg/m², Day 1; repeated on Day 8, 15 and 22 of the first and third course), cyclophosphamide (1 g/m² infused in 1h for 3 doses, Day 1) and etoposide (100 mg/m² infused in 2h, day 1, 2 and 3). In all, 48 completed 3-day chemotherapy courses were administered to 13 patients. The weekly administrations of vincristine after first and third courses amounted to 94 of 130, and 12 of 94 were reduced to 75% of full dose, because of peripheral neurotoxicity. Another 12 of 94 (13%) doses of weekly vincristine were reduced, because of prior severe constipation and peripheral neuropathy. VEC toxicity, as experienced by patients treated in the first AIEOP protocol

Page | 40 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

consisted in: neutropenia grade 4 NCI/CTC after 11 courses, which required precautionary or therapeutic hospitalization in 9 of 11 patients; seven platelet transfusions were required for thrombocytopenia grade 4 in 2 patients, and 27 packed red cell transfusions were given to 8 patients. Inappropriate antidiuretic hormone secretion and post vincristine intestinal ileum complicated the second chemotherapy course in 1 patient. None of the patients suffered toxic death after chemotherapy. This study concluded that VEC chemotherapy could be more widely explored, considering its at least partial efficacy in the small series of patients that were treated. VEC features a substantial lack of severe toxicity and the possibility of rendering a second surgical approach more successful in terms of patient’s morbidity . VEC has been also applied both for patients with post-surgical residual disease and for patients without residual disease but grade 3 tumours (in this case chemotherapy was given after local radiotherapy) also in the 2 AIEOP protocol [97]. Toxicity has not been worse after irradiation. VEC regimen has been applied also in the SIOP ’99 protocol for patients with a residual tumour after surgery. The data from about 38 patients show that 20 obtained complete remission through VEC and second-look surgery ( Richard Grundy, personal communication ). The rationale of adding cisplatin-vincristine to VEC after radiotherapy comes from the fact that these drugs seemed the most active in presence of residual disease in the phase 2 trials so far reported [1]. The Children's Cancer Group protocol 9942 reported on 41 naive patients with residual tumour that were given four cycles of cisplatin-based chemotherapy prior to irradiation. Treatment comprised four cycles of vincristine (1.5 mg/m²/week, 3 doses, max 2 mg/dose, omitted in cycle 4), etoposide (100 mg/m²/day, 3 days), cisplatin (100 mg/m² over 6 hours on the first day of each cycle, with normal saline and mannitol 7.5 g/L pre-hydration), and cyclophosphamide (1000 mg/m² on the second and third days of each cycle, with mesna prophylaxis for hemorrhagic cystitis), repeated every 21 days. Of 35 patients fully evaluable for response to chemotherapy, 14 (40%) demonstrated complete response, 6 (17%) partial response, 10 (29%) minor response or stable disease, and 5 (14%) demonstrated progressive tumour growth. The pre-irradiation chemotherapy group demonstrated EFS comparable to that of patients with no residual tumour who received irradiation alone (55 +/- 8% vs. 58 +/- 9%, P = 0.45). Any benefit of chemotherapy was restricted to patients with greater than 90% tumour resection [43]. Grade 3 or 4 and selected grade 2 toxicities (renal, electrolytes, ototoxicity, and neurotoxicity) were reported in 33 of 41 patients receiving pre-irradiation chemotherapy. These included grades 2–4 hearing changes (17 patients), hypokalemia (13), hypomagnesaemia (10), and peripheral neuropathy (5). Moreover, there is a possibility that dose intensity may be crucial to maintain or improve the response obtained by radiotherapy. Adding drugs that have moderate haematotoxicity such as vincristine and cisplatin to VEC thus seems worthwhile.

1.2.5 Rationale for randomizing high dose methotrexate Methotrexate is an anti-folate metabolite that has been in use for 60 years as an immunosuppressant and antineoplastic agent usually administered once per week. Since the late 1970’s it has been used in very high doses (grams rather than milligrams per square metre) with folinic acid rescue of normal cells until methotrexate levels fall and has an established role in the treatment of osteosarcoma, acute lymphoblastic leukaemia and non-Hodgkin’s lymphoma. Its possible benefit in the management of ependymoma would be in achieving tumouricidal concentrations of methotrexate in both cerebro-spinal fluid and in the brain parenchyma where ependymoma cells may be present.

Page | 41 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

There is a high-level folate receptor expression in ependymoma cells, thereby providing a mechanism to maximise methotrexate accumulation into the tumour [50]. High-dose methotrexate was incorporated into the baby ependymoma United Kingdom Children’s Cancer Study Group CNS 9204 protocol. It has shown the best results published to date with a 5-year radiotherapy-free survival (RTFS) of 42% in patients who did not have metastatic disease at diagnosis [51]. One of the possible reasons for success may be the incorporation of high-dose methotrexate. There are less leptomeningeal CNS metastases in patients treated on this protocol than on other primary chemotherapy-based or radiotherapy-based strategies [52]. The chemotherapy schedule used in the first UKCCSG/SIOP CNS 9204 trial which incorporated high- dose methotrexate into the baby with ependymoma, comprised blocks of alternating myelosuppressive and non myelosuppressive drugs repeated at 14-days intervals after maximal surgical resection. Each course lasted for 56 days and a total of seven cycles were given. Children weighing 10 kg or more were dosed by surface area; those weighing less than 10 kg were dosed by weight as shown. Course 1; carboplatin (550 mg/m² or 20 mg/kg) over 4h and vincristine (1.5 mg/m² or 0.05 mg/kg) intravenous bolus; course 2; methotrexate (8000 mg/m² or 250 mg/kg) and vincristine (1.5 mg/m² or 0.05 mg/kg); course 3; cyclophosphamide (1500 mg/m² or 50 mg/kg) over 4h with prehydration and mesna; course 4; cisplatin (40 mg/m² for 48h or 1.3 mg/kg). 97 children with brain tumours were enrolled. The majority, 70%, experienced grade 4 haematological toxicities, 2 patients experienced grade 3 and 1 patient experienced grade 4 renal toxicities and 1 patient experienced grade 4 ototoxicity. There were no deaths clearly attributable to chemotherapy toxicity. As compared to other similar exclusive chemotherapy protocols, the UKCCSG/SIOP CNS 9204 trial used more intense chemotherapy, and high dose methotrexate. As it provided the best results ever published in baby strategy, the role of high dose methotrexate may be suspected, and deserves a randomised trial. In the HEADSTART I and II trials 5 of 29 patients had metastatic disease and had HD-MTX incorporated into their induction chemotherapy [37]. Three of the 5 had a complete response to induction chemotherapy as opposed to 3 patients with partial response, 6 with stable disease and two with progressive disease out of the 11 patients with measurable local residual disease who did not have treatment with HD-MTX. HD-MTX has never been tested in a stand-alone Phase II study for patients with measurable ependymoma. This present study aims to generate data on methotrexate activity via the window study. It is known that response rates of 40-60% can be obtained for patients with residual ependymoma treated with combination chemotherapy including cyclophosphamide, etoposide and vincristine [53]. It would be unusual for one drug alone to be more effective than 3-4 drugs in combination so it is proposed to test possible response to high-dose methotrexate in a Phase II randomised study of vincristine, cyclophosphamide and etoposide vs vincristine, cyclophosphamide, etoposide and HD-MTX. The proposed trial for patients with residual disease is an elegant design that investigates the possible efficacy of HD-MTX by giving all patients the benefit of VEC chemotherapy whilst randomising half to receive additional HD-MTX. This will generate data that can be used to decide whether methotrexate is worthy of inclusion in future phase III trials. The role of high-dose methotrexate needs to be considered and merits a randomised trial, as this trial provided the best radiotherapy-free results ever published in a strategy for very young children under 3 year of age.

Page | 42 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

1.2.6 Rationale for a boost of radiotherapy The small subgroup of children with residual disease after first or subsequent surgeries represents a challenge in terms of effective adjuvant radiotherapy because of their poor local control with conventional treatments. Radiotherapy using stereotactic technologies represents an attractive method to irradiate, in a single or multiple fractions (radiosurgery or fractionated stereotactic radiotherapy, respectively), radiologically discrete tumour volumes with significant sparing of surrounding normal nervous tissue due to the steep gradient of dose obtained at the margin of the target. At the St. Jude Children’s Research Hospital, five children with incomplete resection were treated with 50.4-55.8 Gy conventional radiotherapy followed by a stereotactic boost, 10-12.5 Gy in a single fraction, to the residual enhancing tumour [54]. Four out of 5 were alive without evidence of progression 14-40 months from radiosurgery. Only 1 child developed an enhancing lesion corresponding to the boosted volume 3 months after treatment that required short-term low-dose corticosteroids and regressed on MRI in 6 months. Stereotactic radiosurgery has been also adopted in the management of recurrent ependymoma with interesting results in terms of 3-year local control of 29% [55] and 68% [56] with acceptable neurological toxicity. In the second AIEOP protocol , opened since 2002, adjuvant radiotherapy implies the delivery of 59.4 Gy, 1.8 Gy per fraction, to the tumour bed with a 3D conformal technique followed (in children with measurable residual disease after first or subsequent surgeries) by a fractionated stereotactic radiotherapy boost, 8 Gy in two 4 Gy fractions. 160 children entered the study. In 24 children out of 40 with residue after first surgery, second look wasn’t feasible or incomplete and thus received VEC and 59.4 Gy to the tumor bed plus 8 Gy to the gross residue. The 5-year estimates for PFS is 58% (95% CI: 39-86) and OS is 69% (95% CI: 50-93) [97]. No iatrogenic death or major toxicity occurred. 4 children irradiated with tomotherapy, developed radiation related MRI changes regressing with steroids within 8 months. Five-year OS and PFS for the whole series were 81% and 65%, respectively. PFS for children with and without residue before radiation therapy was 45% and 72% (p=0.01), and local relapse 32% and 16% (P=0.11), respectively. Thus hypofractionated RT boost was feasible and contributed to obtain local control in children with measurable residue. This procedure will be implemented at a multi institutional level in the current protocol, and should be performed only in those centers that have the skill to do it. Referral to a center with more expertise may be necessary.

1.2.7 Rationale for randomizing valproate in young children receiving front line chemotherapy There is general acceptance that adjuvant therapy is required as part of the treatment for ependymoma in very young children, even when complete resection is achieved [34, 57-61]. Choices over adjuvant therapy have depended on the underlying philosophy of the national group. Radiotherapy is effective but its delivery is complicated by the vulnerability of an immature CNS to radiation damage. Although the degree of functional impairment depends on field size, radiation dose and age at treatment, the majority of long term survivors manifest multiple problems including a global reduction in IQ and more specific cognitive defects such as short term memory loss [7-10]. More recent preliminary studies suggest that conformal radiotherapy to the posterior fossa in infants with ependymoma may not result in such severe neuro-cognitive damage, at least in the short term [3]. However, there are other serious delayed effects from radiotherapy such as neuro- endocrine sequelae and second cancers which may adversely influence quality of life [62, 63]. The perceived unacceptable late effects of cranial radiotherapy in young children led a number of

Page | 43 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

institutions and national groups to the adoption of chemotherapy based strategies designed to avoid or delay irradiation [57, 58, 61, 64]. The first approach was used in the UK and French Children’s Oncology groups (CCLG 9204 protocol and the SFOP study) These studies have produced some of the best survival and radiotherapy-free survival figures published for children with this condition (table 1).This first UKCCSG infant brain study (CNS9204) differed from other studies in the rapid scheduling of agents which alternated myelosuppressive with less myelosuppressive chemotherapy [65]. Furthermore, patients were only irradiated at the time of disease progression or at the time of subsequent relapse. International consensus for children less than 12 months of age favours the use of a primary chemotherapy regimen. Importantly National preferences alongside due consideration of anatomical location of tumour and parental views will be taken into account. The SFOP study used post-operative chemotherapy consisting of 7 cycles alternating two drugs for each course over a year and a half without irradiation in children up to 5 years old: 73 children were enrolled into this multicentre trial. With a median follow-up of 4.7 years, the 4-year OS was 59% and the PFS were 22%; 40% were radiation-free at 2 year [57]. The randomized controlled Pediatric Oncology Group study 9233 tested the hypothesis that dose- intensive chemotherapy would improve event-free survival (EFS) for children <3 years of age with different newly diagnosed malignant brain tumors. Dose-intensive chemotherapy resulted in significantly improved EFS distribution (P = .0011) (2-year EFS rates of 42.1% vs. 19.6% with SD chemotherapy), but not OS distribution, for patients with centrally confirmed ependymoma [67]. The CCLG CNS 9204 study used an intensive chemotherapy alternating myelosuppressive and non myelosuppressive drugs each every 2 weeks for 1 year. It accrued 89 infants, < 3 years, with ependymoma. In addition 97 children with other brain tumours were enrolled [98]. Chemotherapy schedule comprised blocks of alternating myelosuppressive and non myelosuppressive drugs repeated at 14-days intervals. With a median follow-up of 6 years, the OS for the 80 non-metastatic patients at 3 and 5 years was respectively 79.3% and 63.4% and EFS was 47.6% and 41.8%; 42% were radiotherapy-free at 5 years. Local relapse occurred in 47 of 59 (80%) patients who progressed. The highest relative dose intensity of chemotherapy increased the 5-year overall survival to 76%, compared with 52% for the lowest relative dose intensity. The results for infants with incompletely resected ependymoma compare favourably with other published results, but are still far from satisfactory [51]. A meta-analysis of primary chemotherapy strategies was undertaken and revealed that the CCLG CNS 9204 study provided a superior outcome (Richard Grundy Personal communication). There is little data available on infants below the age 12 months with an initial diagnosis of intracranial ependymoma. In this study the standard arm of chemotherapy will be composed of the CCLG CNS 9204 drug /cycles and the experimental arm will be the CCLG CNS 9204 chemotherapy schedule combined with valproate used as a histone deacetylase inhibitor (HDACi). HDAC are known to regulate gene transcription and oncogenesis through remodelling of chromatin structure, canonically resulting in the repression of target genes [66]. Small-molecule histone deacetylase inhibitors (HDACi) have achieved significant biological effects in preclinical cancer models, including brain tumours such as glioblastoma multiforme (GBM) [67] and medulloblastoma [68]. Trichostatin-A (TSA), a potent HDACi caused a dramatic reduction in cell proliferation, activated a DNA damage response, altered cell cycle distribution, and induced caspase-3–dependent apoptosis

Page | 44 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

in high-grade childhood CNS PNET, medulloblastoma, and ependymoma [69]. More recently, vorinostat has been shown to have differentiation inducing potential in an independent ependymoma cell line [70]. Concerns that HDACi may interfere with critical cellular functions seem unfounded as evidence thus far indicates that HDACi display selective toxicity against tumour cells compared with untransformed cells [71]. Furthermore, TSA did not disrupt normal cellular function in a rodent ependymal primary cell culture model [69]. HDACi equally sensitize cancer cells to the cytotoxic effects of other chemotherapeutic agents and radiotherapy [72]. To a variable extent, HDACi induce growth arrest, differentiation, or apoptosis in vitro and in vivo. All of the above suggest that inhibitors of histone deacetylase are potentially effective anticancer agents.

Presently, very little is known about the mechanism of action of HDACi as prior studies of HDACi in brain tumour cells did not investigate the mode of HDACi action or the temporal order of cellular events post HDACi exposure. The mechanism(s) by which HDACi exert downstream effects that attenuate tumour growth is unknown. One attractive candidate for HDACi-mediated inhibition is the holoenzyme telomerase. In 90% of malignant cells, unlimited replicative potential is conferred in considerable part by the maintenance of telomere length at chromosomal termini, above a critical minimum required to bypass senescence and apoptosis. This is achieved through the de novo addition of telomeric repeats by telomerase, which consists of an intrinsic RNA template (human telomerase RNA, hTR) and a catalytic component [human telomerase reverse transcriptase (hTERT)]. High telomerase activity is evident in 90% of cancers tested, whereas in most somatic tissues the enzyme is undetectable or transiently active at low levels due to repression of hTERT transcription. Telomerase has been detected in an increasing number of brain tumours, including those arising in childhood. Whether these anti-telomerase effects are consequential or causal to the observed growth arrest remains unclear, as is the mechanism of TSA-mediated telomerase inhibition. Previously published preclinical studies and clinical trials in paediatric patient populations have indicated a clear rationale for the targeting of trough valproate levels, with a view to increasing the likelihood of antitumour activity [73].

Promising in vitro results has led to several clinical trials in brain tumours with diverse HDACi. Two main HDACi have been used: vorinostat and valproate has been used as a HDACi in a number of clinical trials and has been found to be safe and tolerable even in heavily pre-treated patients [74]. A recent phase I study conducted with Children’s Oncology Group has defined an appropriate dosing schedule [75]. An optimal dosing strategy of three times daily administration, with a starting dose of 15 mg/kg/day and subsequent weekly dose escalations, to achieve trough plasma VPA concentrations of 75-100 µg/ml was advised. This dosing regimen shortened the time required to achieve the desired VPA concentrations and was not associated with dose limiting toxicities. However, further work is required to define optimal anticancer valproate concentrations and identify appropriate pharmacodynamic biomarkers of activity and response.

The selected dose is based on the papers from Su et al [75] and Woolf et al [76]. However, a twice daily dosing schedule has been chosen to make this easier for the parents and the children. Given the long half-life of valproate, this dosing schedule should not cause toxicities but will be carefully evaluated as part of a pharmacokinetic study. For the serum levels of VPA should be around 100 – 150 micrograms per ml. Previous studies by Woolf et al and Su et al have shown that trough concentrations of around 100 micrograms per ml were well tolerated in children with refractory CNS tumours and in heavily pre-treated paediatric patients [76]. As patients are presenting with newly

Page | 45 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

diagnosed disease, the serum trough levels can be increased slightly to maximise the efficacy of histone deacetylase inhibition. Data from Professor Olaf Whitt (personal communication) shows the need to maintain levels between 100 and 150 micrograms per ml to obtain efficacious anti-cancer activity. Furthermore it concluded that valproate deserves further studies as a novel agent in paediatric cancers, especially in combination with radiation treatment, chemotherapeutic drugs or biologic agents in children with CNS tumours. Further studies suggest that the addition of VPA improves outcomes in paediatric high grade glioma (pHGG). Furthermore, a recent retrospective analysis of the outcome of patients with high grade glioma who were taking VPA as an anti-epileptic medication had an improved survival compared to those that were not when matched for clinical variables known to affect outcome in HGG [76] . The schedule of the chemotherapy strategy for very young children with ependymoma is dose intense with patients receiving treatment every 14 days. Patients will undergo regular monitoring of their full blood count, liver function tests and dose modifications will be introduced for Grade IV toxicities. Patients will be supported by the cytokine GCSF to maintain blood counts within the normal range. This study will therefore seek to determine whether VPA (valproate), as a HDACi, will improve the PFS of children ineligible to receive radiotherapy as part of a randomised clinical trial. A pharmacokinetic study of valproate therapy will be undertaken and correlated with clinical outcomes. Available evidence suggests that the addition of VPA is unlikely to cause adverse risk and toxicity. The potential benefit of enhanced cytotoxicity and improved outcomes therefore outweigh risk of VPA.

1.2.8 Rationale to evaluate neuropsychological and neurocognitive outcomes Survival is often associated with an impaired quality of life. In the past decades, studies on the sequelae of brain tumours in children and adolescents focused on motor, sensory, cognitive and neuropsychological sequelae as well as on the actual psychological and behavioural modifications (although to a lesser extent) that impact negatively on re-entry to school and community. Studies on the cognitive outcome of patients with ependymoma are limited in number and are generally part of larger studies that include patients with other posterior fossa tumours such as medulloblastomas and astrocytomas. The most recent studies on cognitive and neuropsychological outcomes show that patients treated for ependymoma do not present with severe cognitive impairments when compared to other types of brain tumour patients, even though many of them eventually develop a broad array of neuropsychological problems that impair formal learning and quality of life [33, 77-80]. In particular, patients with posterior fossa tumours are considered to be at an especially higher risk of neuropsychological sequelae and such tumours have been investigated by neuropsychological studies more than tumours affecting other brain areas [81]. Age at diagnosis and treatment, (longer) time since treatment, the aggressiveness of surgery and the extent of radiotherapy are considered to be important predictive factors for the onset of cognitive and neuropsychological problems which have a great impact on the quality of long-term survival. Throughout the ependymoma program, all the patients enrolled into interventional strata will receive an age-appropriate clinical neurocognitive evaluation.

Page | 46 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

1.2.9 Rationale for registering metastatic and relapsing patients The incidence of metastases reported in the literature is low in children after infancy. However, this may not have been systematically explored in the past [82, 83]. As in other diseases, it is an hypothesis that metastatic patients have specific molecular features [84]. It is thus crucial to evaluate for these patients the key molecular events. One aim of this protocol is to obtain tumour sample and cerebrospinal fluid (CSF) from all patients. If the patient is metastatic no specific treatment guidelines are available, but national coordinators should be contacted for advice. Similarly, patients who relapse may provide crucial information, if their tumour can be collected and compared to the initial sample. Here again, different advice may be provided by the national trial coordinators, concerning the possibility of second intervention, re-irradiation, or participation in a currently on-going phase II studies in this disease [3, 85, 86].

1.2.10 Rationale for registering spinal ependymomas Primary spinal cord tumours are very rare in this age group with an overall incidence rate of 0.26 per 100.000 person-years [87]. Although the majority of “adult” studies included small numbers of children, only few retrospective studies on children with spinal cord ependymoma have been performed. Recently the HIT study group presented a series including 29 children [88]. With a median follow-up of 4.2 years 28 patients (96.6%) were alive and 7 patients (24.1%) developed progressive disease or relapsed. Four of these 7 progressive /relapsing patients were alive at last follow-up without evidence of disease (n=1, ependymoma) or with stable residual lesions (n=3, myxopapillary ependymoma) following repeated surgery ± local radiotherapy. Two patients were alive with progressive disease (myxopapillary ependymoma, n=1; ependymoma, n=1) and one patient died of progressive disease (anaplastic ependymoma), all of them having received repeated salvage therapies. Overall and progression-free survival rates at 5 years in this group were comparable to those reported by others. None of the parameters tested in the HIT cohort (age, gender, site of the tumour, grading) had an impact on survival. The number of patients that had a GTR (58.6%) in the HIT study is in the range reported for adult patients. There was a trend towards improved PFS among patients with GTR when compared to patients with incomplete resection (PFS 84.4% vs. 57.1%, p=0.088). This encourages a prospective collection of data at a European level .

1.2.11 Benefit and Risk assessments Standard multimodal therapeutic options include surgery, radiotherapy and chemotherapy and depend on the age of the patient and the location of the primary site of the tumour. The aim of the SIOP Ependymoma II Programme is to improve the overall survival of patient with ependymoma by optimizing the standard of care of patient following initial surgery. This program explores chemotherapies which are already studied through a few trials or cohorts which are too small to conclude on the balance benefit /risks of the intervention and to provide a new recommendation on standard of care. The introduction of new chemotherapy practices can only be considered through large randomized clinical trials as the SIOP Ependymoma II Program. The chemotherapy selected in experimental plans proposed by SIOP Ependymoma II Program is based on an extensive review of the previously published clinical trials and a constant exchange of experience throughout the worldwide paediatric oncology community through the international Society of Paediatric Oncology platform. The use and the safety of the regimens proposed in the SIOP Ependymoma II Program are supported by the followings:

Page | 47 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

• VEC and VEC+CDDP are all part of the therapeutic options for patients with intracranial ependymoma and have been largely used, and are currently used, in several clinical trials in paediatric patients with intracranial ependymoma. Previous trials with the Investigational Medicinal Products (IMPs) used in ependymoma provided encouraging results and both regimens have demonstrated a cytotoxic activity and can be regarded as standard regimens that are applied in currently open, paediatric, ependymoma protocols. Most of these IMPs are already used routinely in European participating countries however they do not have Market authorization in this indication.

• IMPs will be administered at the recommended therapeutic daily dose in SmPC (other indications) and in previous clinical trials described above. The dose regimens and the schedules of administration proposed in the ependymoma II program are similar to those used in previously reported protocols. Although, the use of such dose chemotherapy regimens may generate additional toxicity, the sponsor considers that the potential improvement of PFS and OS justifies the use of these treatments.

For stratum 1 and 2: o Several paediatric studies have demonstrated that dose regimen chosen in the ependymoma were associated with acceptable toxicity (see above). o Specific recommendations for the management of treatment-related toxicity regarding vincristine, etoposide, cyclophosphamide and methotrexate toxicities are precisely defined in the study protocol: ° Haematological toxicity: Courses of chemotherapy should be delayed until neutrophils of at least 1.0 x 10 9/l and platelets 100 x 10 9/l. ° Ototoxicity: Audiometry should be performed prior to first and third CDDP courses of chemotherapy. ° Nephrotoxicity: Both glomerular and tubular toxicity must be monitored during treatment. The estimation of GFR must be performed before the first and third course. ° Bladder toxicity: Adequate infusion with hydration and mesna is specified for each IMP. ° The supportive care during chemotherapy is stated including anti-emetic treatment, antibiotics for suspected fever, neutropenia and treatment with G-CSF for infection. ° The main risks for patients are the side-effects related to IMPS. Most of which can be managed with prophylactic or therapeutic treatments. ° For methotrexate, specific dose modification has been defined in cases of mucositis, drug interactions, or nephrotoxicity and delayed excretion of methotrexate.

For stratum 3: o The CCLG9204 study used intensive chemotherapy alternating myelosuppressive and non myelosuppressive drugs delivered every 14 days weeks over 1 year and has shown acceptable outcome but there is clearly room for improvement. Valproate could be the anticancer agent that allows this improvement, since recent studies using Valproate

Page | 48 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

(described above) suggested that the addition of this drug improves outcomes of paediatric patients with high grade glioma. Despite the heavily pre-treated patient population in this reported study, the toxicity of valproate was very mild. There was no toxic death. Minor toxic reactions were reported but there was no detectable relationship between valproate level and outcome. o Specific recommendations for the management of treatment-related toxicity have been also considered in this stratum and patients will undergo regular monitoring of their full blood count, liver function tests and dose modifications will be introduced for Grade IV toxicities. Patients will be supported by the cytokine GCSF to maintain blood counts within the normal range. o Given the long half-life of valproate, this dosing schedule should not cause toxicities but will be carefully evaluated as part of a pharmacokinetic study which of valproate therapy will be correlated with clinical outcomes. Available evidence suggests that the addition of valproate is unlikely to cause adverse risk and toxicity. The potential benefit of enhanced cytotoxicity and improved outcomes therefore outweigh risk of valproate.

The proposed experimental plan and the specific recommendations for treatment infusion and management of treatment-related toxicity justify the use and the safety of the chemotherapeutic regimens proposed in this trial. We believe that the potential benefits of the therapies being investigated in this program outweigh the risks associated with study participation. The doses used in the ependymoma program are based on analysis of data from the literature described above as well as the dosage recommendations from Summaries of Product Characteristics. A toxicity monitoring plan has been set up for each stratum: guidelines on the monitoring toxicities chemotherapies and modification of the doses of chemotherapies have been considered in this protocol. Additional toxicity of chemotherapies requiring greater support should be manageable by following the protocol recommendations. However, improvement of PFS and OS might justify the consideration of the therapeutic strategies proposed.

In conclusion, based on previous clinical experience, the proposed study design, the dose selection, and the safety monitoring plan described in this protocol, the Sponsor considers that adequate risk mitigation measures are included in this protocol and that the benefit-risk profile is favourable to proceed with the proposed clinical study .

Page | 49 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

2 AIMS, OBJECTIVES AND OUTCOME MEASURES

2.1 Aims and objectives

2.1.1 Overall program The aim of the overall program is to improve the clinical outcome of patients with ependymoma. It will be obtained through randomised clinical trial optimising the standard of care following initial surgery. Patients not included in interventional arms will be registered in the observational study to improve our knowledge of the natural history of this disease. An integrated study will be conducted with the BIOMECA consortium on tumour material from patients to identify prognostic and predictive biomarkers.

The primary objective is to determine whether the assessment of residual disease can be improved by a centralized review of post-operative MRI and whether such review increases the rate of complete resection compared to historical controls. Does central neurosurgical and radiological review increase resection rates?

Secondary objectives: • To evaluate second look surgery rates as compared to historical controls.

Exploratory objective (integrated BIOMECA study): • To test the hypothesis that key molecular events are predictive of clinical behaviour of ependymoma. • To prospectively evaluate 1q copy-number status, Tenascin C, RELA-fusion, YAP fusion, H3.3K27me3 and molecular subgroup (by methylation array) as prognostic and predictive biomarkers in ependymoma within clinical trial setting.

This part of the program will be organised by the European Ependymoma biology consortium called “Biomarkers of Ependymomas in Children and Adolescents (BIOMECA)” throughout a cooperation agreement with the aim to identify informative prognostic biomarkers for assessment of disease status and predictive response to therapy:

• to identify new biomarkers for Ependymoma clinical and biological behaviour (location, recurrence, chemo resistance, invasion, metastasis). • to validate known biomarkers in the context of the new international prospective study for newly diagnosed Ependymoma • to provide a new comprehensive grading scheme based on the histological, and immunohistochemical and/or biological criteria that could be used for newly diagnosed Ependymoma. • to validate and compare the techniques to assess biomarkers for further use in the stratification of patients • to select the best biomarker(s) and establish a prognostic signature for ependymoma.

Page | 50 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

2.1.2 Treatment stratum 1: Patients with no measurable residual disease and ≥ 12 months of age (phase III) The primary objective is to compare PFS in patients who receive 16 weeks of chemotherapy with VEC+CDDP following complete surgical resection and radiotherapy to those who undergo complete surgical resection and radiotherapy alone.

The secondary objectives are: • to determine whether OS is improved in patients who will receive 16 weeks of chemotherapy (VEC+CDDP) following surgical resection and radiotherapy when compared to OS of patients who will undergo surgical resection and radiotherapy alone, • to compare the neuroendocrine morbidity in each treatment arm, • to evaluate the neuropsychological morbidity in each treatment arm, • to evaluate the QoS in each treatment arm, • to determine the safety and the tolerance to 16 weeks of chemotherapy (VEC+CDDP) following surgical resection and radiotherapy when compared to those that undergo surgical resection and radiotherapy alone.

2.1.3 Treatment stratum 2: Patients with inoperable measurable residue and ≥ 12 months of age (phase II) The primary objective is to compare the activity of 2 postoperative chemotherapy schedules with VEC or VEC+HD-MTX in patients who have incompletely resected tumours.

The secondary objectives are: • to determine the safety and tolerability to chemotherapy VEC+HD-MTX following surgical resection when compared to those who receive VEC alone, • to evaluate whether OS is improved in patients with VEC+HD-MTX following surgical resection when compared to those who receive VEC alone, • to evaluate whether PFS is improved in patients with VEC+HD-MTX Dose following surgical resection when compared to those who receive VEC alone, • to compare the neuroendocrine morbidity in each treatment arm. • to evaluate the neuropsychological morbidity in each treatment arm • to evaluate the QoS in each treatment arm.

Exploratory objective: is to determine safety of 8 Gy boost radiotherapy in patients with residual disease despite frontline chemotherapy and after a 59 Gy conformal radiotherapy .

2.1.4 Treatment stratum 3: Randomised phase II chemotherapy study in children < 12 months of age or those not eligible to receive radiotherapy (phase II) The primary objective is to evaluate the activity of adding valproate to a primary chemotherapy for children unable to receive radiation therapy. To evaluate the progression free survival in children unable to receive radiation therapy and who receive valproate, as a histone deacetylase inhibitor in addition to the primary chemotherapy strategy when compared to those that undergo chemotherapy without valproate.

Page | 51 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Secondary objectives are • to evaluate whether OS is improved in patients who receive valproate added to a primary chemotherapy strategy when compared to those that receive primary chemotherapy only. • to evaluate whether radiotherapy- and progression- free survival are improved in patients who receive valproate added to a primary chemotherapy when compared to those who receive primary chemotherapy only. • to compare the neuroendocrine morbidity in each treatment arm, • to evaluate the neuropsychological morbidity in each treatment arm, • to evaluate the QoS in each treatment arm, • to determine the safety and tolerability of valproate added to a primary chemotherapy strategy for children unable to receive radiation therapy.

2.1.5 Observational study After staging, patients that will not fulfil the inclusion criteria of one of the interventional studies will be enrolled and followed up into an observational study for which only descriptive analyses will be performed.

2.2 Outcome measures

2.2.1 Overall program Primary endpoint measure: Gross Total Resection (GTR) rate

Secondary endpoints measure: Second look surgery rate

Only descriptive statistics will be produced for GTR rate and second look surgery rate.

2.2.2 Treatment stratum 1: Patients with no measurable residual disease and ≥ 12 months of age (phase III) Primary endpoint measure: PFS from the date of randomisation to the date of event defined as progression or death due to any cause.

Secondary endpoints measures: Overall survival measured from the date of randomisation to the date of death due to any cause. Quality of survival (QoS) Neuropsychological outcomes Neuroendocrine outcomes (neuroendocrine late effects) Short and long-term safety: Adverse Events (CTCAE v4.03 )

2.2.3 Treatment stratum 2: Patients with inoperable measurable residue and ≥ 12 months of age (phase II) Primary endpoint measure: Number of treatment responders. Objective response to chemotherapy is measured based on SIOP-E Neuro Imaging guidelines.

Page | 52 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Secondary endpoint measures: OS measured from the date of randomisation to the date of death due to any cause, PFS from the date of randomisation to the date of event defined as progression or death due to any cause, Quality of survival (QoS) Neuropsychological outcomes Neuroendocrine outcomes (neuroendocrine late effects) Short and long-term safety of frontline chemotherapy: Adverse Events (CTCAE v4.03 )

Exploratory endpoint measure: Event free survival for patients with boost of radiotherapy. Toxicity will be monitored in the subgroup receiving radiotherapy boost.

2.2.4 Treatment stratum 3: Randomized phase II chemotherapy study in children <12 months of age or those not eligible to receive radiotherapy (phase II) Primary endpoint measure: PFS from the date of randomisation to the date of event defined as progression or death due to any cause.

Secondary endpoint measures: OS measured from the date of randomisation to the date of death due to any cause. Radiotherapy free survival rate Quality of survival (QoS) Neuropsychological outcomes Neuroendocrine outcomes (neuroendocrine late effects) Short and long-term tolerance profile of frontline chemotherapy based on proportion of patients experiencing toxicity grade 3 to 4 adverse events (CTCAE v4.03 ).

Exploratory Endpoint measures (optional): Pharmacokinetic modelling will be carried out using valproate pharmacokinetic parameters in conjunction with patient characteristics and clinical parameters in order to investigate the key factors involved in determining individual valproate drug exposures within the patient population. Valproate pharmacodynamics will be followed throughout changes in histone H3 and H4 acetylation. Changes between baseline and time of steady state valproate will be correlated with valproate trough levels and clinical response.

Page | 53 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

3 TRIAL DESIGN

This international multi-centre clinical program will enrol all patients with a primary diagnosis of ependymoma WHO grades I, II and III. This program includes a staging phase where patients will be screened for eligibility prior to patients’ specific care. Each case will be carefully reviewed at national level by cancer specialists from multidisciplinary teams. This review will include an evaluation of postoperative report, neuroimaging and pathology reports to confirm the diagnosis of ependymoma, the grading and the status of the disease. In a second step, patients will be offered the opportunity to take part in one of the three interventional clinical strata with therapeutic strategies adapted to the disease status (residual vs no residual disease), to age (< 12 months vs ≥ 12 months) and according to national policy. Patients who are not eligible for the interventional studies will be followed up in an observational study that is also part of the global program. Each clinical stratum of the ependymoma program has been designed taking into account the rarity of the disease and the most appropriate statistical assumptions. • Stratum 1 – Patients ≥ 12 months of age at diagnosis and with no measurable residual disease will be enrolled in a randomised trial of 16 weeks of chemotherapy with VEC- Cisplatin versus observation only, after the completion of radiotherapy. On confirmation of entry eligibility to this stratum, all patients will be immediately randomised to one of 2 arms: conformal radiotherapy or conformal radiotherapy plus chemotherapy. Conformal RT (cRT) will be performed for all patients enrolled in this stratum. They will receive cRT of 59.4 Gy (only 54 Gy if < 18 months or with risk factors as multiple surgeries (more than 2) or poor neurological status) with a 0.5 cm margin in daily 1.8 Gy fractions (under general anaesthesia if required), 5 fractions per week. Conformal radiotherapy will be followed either by 16-weeks of chemotherapy with VEC+CDDP (started 4 weeks after completion of radiotherapy) or by observation.

• Stratum 2 – Patients ≥ 12 months at diagnosis with confirmed inoperable measurable residue will be enrolled in randomized frontline phase II chemotherapy study and exploration of the efficacy of a boost of radiotherapy On confirmation of entry eligibility to this stratum, all patients will be immediately randomised to one of 2 arms: they will receive an 8-week course of postoperative induction chemotherapy comprising VEC+HD-MTX or VEC. Chemotherapy should be started as soon as recovered from surgery and ideally within 3 weeks of surgery. Then, all patients will be evaluated by MRI for a second look surgery whenever possible; all patients will receive cRT to the tumour bed. In case of persisting residual tumour after cRT, a boost of 8 Gy to the residual tumour (2 fractions of 4 Gy on 2 consecutive treatment days) will be considered. All patients will then receive a 16-week maintenance chemotherapy with VEC+CDDP provided they were not progressive under previous chemotherapy.

Page | 54 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

• Stratum 3 – Children < 12 months or those not eligible to receive radiotherapy will be enrolled in a randomised phase II chemotherapy study including dose intense chemotherapy alone or in association with valproate. On confirmation of entry eligibility to this stratum, patients will be randomised to receive 2 different schedules of treatment, with valproate or not. Each cycle of chemotherapy comprises 4 courses of alternating myelosuppressive and relatively non-myelosuppressive drugs repeated every 56 days for a total of 7 cycles. The randomisation determines if the patient has additional valproate or not. At the end of chemotherapy, patients who have received valproate from the beginning of chemotherapy will stay on drug for an additional year unless they progress. Chemotherapy should be started as soon as the patient has recovered from their operation and ideally within 3 weeks of surgery.

• Observational study After staging, patients who are not eligible to any of the 3 interventional studies or who refuse to comply with protocol specifications of the interventional study proposed will be registered and followed up according to national guidelines.

3.1 Initial local procedure In case of suspicion of ependymoma based on imaging and frozen smear evaluation whilst the patient is undergoing surgery, we recommend the following procedures: • Local clinician to contact national ependymoma panel of potential new case • Local clinician needs to liaise with the neurosurgical team to ensure post-operative MRI scan is done within 72 hours if not done intraoperatively

3.2 National Central Radiological Review Within 2 weeks following MRI, the imaging data must be submitted for central review in DICOM format on a CD or electronically.

• If a systematic centralized radiological review is already implemented as a routine diagnostics (irrespective of study inclusion) in the participating country concerned, the MRI imaging review will follow the usual procedure used in the country concerned in the respect of national laws in force, namely respecting confidentiality of patient’s personal details. • • If no systematic centralized radiological review is already implemented in the participating country concerned, the procedure to be followed is: ° According to ICH and national regulation in force, the investigational centre makes anonymized copy of: − Preoperative and postoperative MRI pictures on CD − Local radiological report specifying the parameters used, deviations measured and the local conclusions − Surgery report

Page | 55 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

° Then, the investigational centre will forward the anonymized preoperative and postoperative MRI pictures, local radiological report and surgery report to the national reference pediatric neuro radiologist for review ° Film copies may be accepted only with the approval of the national reference radiologist

3.3 National Central pathological review Priority is to ensure that sufficient material is available for the neuro pathologist to make a primary diagnosis. The neurosurgeon should aim to ensure that maximal amount of tissue is sent to the pathology department for further processing. Pathologists are responsible for assuring that sufficient interpretable material will be available for further biological work which is an integral part of this study. ° If a systematic centralized pathological review is already implemented as a routine diagnostics (irrespective of study inclusion) in the participating country concerned, the pathological review will follow the usual procedure used in the country concerned in the respect of national laws in force, namely respecting confidentiality of patient’s personal details. ° If no systematic centralized pathological review is already implemented in the participating country concerned, the procedure to be followed is: An anonymized local pathological review report will also be submitted to the national reference pathologist for central pathological review. The submission of pathology reports shall be made directly from the participating institutions.

The reference pathologist will confirm the histological diagnosis. The classification and the grading of the tumours will be performed according to WHO criteria and to the new grading system. The Reference pathologist or any other qualified person (identified and authorized prior any records), send the central pathological report to: ° The investigational centre (The principal investigator of the centre) ° The national investigator coordinator of the country concerned ° The national coordinating centre of the country concerned According to national practices, results from central pathological review will be entered into the eCRF either by the national referent pathologist or by the investigational centre.

Institutions must submit materials for central review after initial surgery.

Page | 56 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

4 ELIGIBILITY

All clinical and laboratory data required for initial work up of any included patient must be available in the patient’s medical record which will serve as the source document for verification at the time of monitoring and/or audit.

4.1 Staging Phase

4.1.1 Inclusion criteria • Main residence in one of the participating countries • Age < 22 years old at the diagnosis • Histological diagnosis of intracranial or spinal, localized or metastatic, ependymoma according to local pathologist (all WHO grades) including: myxopapillary ependymoma, ependymoma (papillary, clear-cell, tanycytic), ependymoma RELA fusion-positive or anaplastic ependymoma, • Delivery to national referral pathology center of FFPE tumour tissue blocks (or at least twenty 5 µm sections on charged slides with sufficient interpretable material and at least ten 10 µm curls in an Eppendorf tube), • Written informed consent (specific to staging) for data and study biological samples collection, • All patients and/or their parents or legal guardians willing and able to comply with protocol schedule and agree to sign a written informed consent, • Patients must be affiliated to a Social Security System in countries where this is mandatory.

4.1.2 Exclusion criteria • Patient with subependymomas and ependymoblastomas, • Primary diagnosis predating the activation of the SIOP Ependymoma II program (Apr 29 th 2015).

4.2 STRATUM 1

4.2.1 Inclusion criteria • Age > 12 months and < 22 years at time of study entry, • No residual measurable ependymoma based on the central neuro-radiological review ( please refer to radiological assessment: section Definitions of residual tumour ) This includes: ‹ R0: No residual tumour on postoperative MRI in accordance with the neurosurgical report ‹ R1 : No residual tumour on MRI but description of a small residual tumour by the neurosurgeon. ‹ R2 : small residual tumour on MRI with the maximum diameter below 5mm in any direction • Newly diagnosed intracranial ependymoma of WHO grade II-III confirmed by central pathological review, • No metastasis on spinal MRI and on CSF cytology assessments ( see section Mandatory Lumbar puncture ), • No previous radiotherapy, • No previous chemotherapy (except steroids), • No co-existent unrelated disease (e.g. renal, hematological) at the time of study entry that would render the patient unable to receive chemotherapy, • No medical contraindication to radiotherapy, and chemotherapy,

Page | 57 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

• No signs of infection, • Adequate bone marrow function: o WBC > 2 x10 9/l o Peripheral absolute neutrophil count (ANC) ≥ 1 x 109/l o Platelet count ≥ 100 x 10 9/l, Hemoglobin > 8.0 g/dl (may have PRBC transfusion to achieve Hb > 8g/dl) • Adequate liver function: o Aspartate aminotransferase (AST) and/or alanine aminotransferase (ALT) ≤ 3 x institutional upper limit of normal for age o Total bilirubin ≤ 2 x institutional upper limit of normal for age • Adequate renal function: < 1.5 x Normal serum creatinine as adjusted for age: Age Maximum serum creatinine (mg/dl) < 5 years 0.8 > 5 < 10 years 1.0 > 10 < 15 years 1.2 > 15 years 1.5 Age Maximum serum creatinine (µmol/l) < 5 years 46 > 5 < 10 years 51 > 10 < 15 years 60 male, 55 female > 15 years 80 male, 70 female In case of suspected compromised renal function, glomerular filtration should be measured by an isotope GFR method such as EDTA clearance or by creatinine clearance. GFR should be: o > 80 ml/min/1.73m² for children 2 years and over, o > 70 ml/min/1.73 m² for children aged 18-23 months, o >65 ml/min/1.73 m² for children aged 12-17 months, • Post-menarchal female not pregnant or nursing (breast feeding) and with a negative beta-HCG pregnancy test prior to commencing the trial, • Males and females of reproductive age and childbearing potential with effective contraception (see Appendix 7 - Definition of highly effective methods of contraception ) for the duration of their treatment and 6 months after the completion of their treatment, • Patients and/or their parents or legal guardians must be willing and able to comply with scheduled visits, treatment plan, laboratory tests and other study procedures of the stratum 1, • Written informed consent (specific to stratum 1) from patients and/or their parents or legal guardian(s) obtained for study participation, data acquisition and transfer, collection and transfer of biological samples. Assent, of the child, when appropriate, will be obtained according to institutional guidelines, • Patients must be affiliated to a Social Security System in countries where this is mandatory.

4.2.2 Exclusion criteria • Tumour entity other than primary intracranial ependymoma, • Patients with WHO grade I ependymoma including myxopapillary variant, • Patients with spinal cord location of the primary tumour, • Participation within a different trial for treatment of ependymoma,

Page | 58 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

• Concurrent treatment with any anti-tumour agents, • Inability to tolerate chemotherapy, • Unable to tolerate intravenous hydration, • Other severe acute or chronic medical or psychiatric conditions or laboratory abnormalities that may increase the risk associated with study participation or investigational product administration, or may interfere with the interpretation of study results in the judgment of the investigator, • Pre-existing mucositis, peptic ulcer, inflammatory bowel disease, ascites, or pleural effusion, • Contraindication to one of the IMP used in the stratum 1 according to the SmPCs in appendix 4 of this protocol ( SmPCs in appendix are those from UK which were chosen for the assessment of the safety aspects of the study ), • Patient for whom imaging remains RX despite all effort to clarify the MRI conclusion (see section Definitions of residual tumour) .

4.3 STRATUM 2

4.3.1 Inclusion criteria • Age > 12 months and < 22 years at time of study entry, • Residual non reoperable measurable ependymoma based on central neuro-radiological review (R3-4 residual tumour that can be measured in three planes, for definition of residuum please refer to radiological assessment : section Definitions of residual tumour) : ‹ R3: Residual tumour that can be measured in 3 planes, ‹ R4: Size of the residual tumour not differing from the preoperative status (e.g. after biopsy) • Newly diagnosed intracranial ependymoma of WHO grade II-III confirmed by central pathological review, • No metastasis on spinal MRI and on CSF cytology assessments (see section Lumbar puncture ), • No previous radiotherapy, • No previous chemotherapy (except steroids), • No co-existent unrelated disease (e.g. renal, hematological) at the time of study entry that would render the patient unable to receive chemotherapy, • No medical contraindication to radiotherapy, and chemotherapy, • No signs of infection, • Adequate bone marrow function: o WBC > 2 x10 9/l o Peripheral absolute neutrophil count (ANC) ≥ 1 x 109/l o Platelet count ≥ 100 x 10 9/l, Hemoglobin > 8.0 g/dl (may have PRBC transfusion to achieve Hb > 8g/dl) • Adequate liver function: o Aspartate aminotransferase (AST) and/or alanine aminotransferase (ALT) ≤ 3 x institutional upper limit of normal for age o Total bilirubin ≤ 2 x institutional upper limit of normal for age • Adequate renal function: < 1.5 x Normal serum creatinine as adjusted for age: Age Maximum serum creatinine (mg/dl) < 5 years 0.8 > 5 < 10 years 1.0

Page | 59 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

> 10 < 15 years 1.2 > 15 years 1.5 Age Maximum serum creatinine (µmol/l) < 5 years 46 > 5 < 10 years 51 > 10 < 15 years 60 male, 55 female > 15 years 80 male, 70 female In case of suspected compromised renal function, glomerular filtration should be measured by an isotope GFR method such as EDTA clearance or by creatinine clearance. GFR should be: o > 80 ml/min/1.73m² for children 2 years and over, o > 70 ml/min/1.73 m² for children aged 18-23 months, o >65 ml/min/1.73 m² for children aged 12-17 months, • Post-menarchal female not pregnant or nursing (breast feeding) and with a negative beta-HCG pregnancy test prior to commencing the trial, • Males and females of reproductive age and childbearing potential with effective contraception (see Appendix 7 - Definition of highly effective methods of contraception ) for the duration of their treatment and 6 months after the completion of their treatment, • Patients and/or their parents or legal guardians must be willing and able to comply with scheduled visits, treatment plan, laboratory tests and other study procedures of the stratum 2, • Written informed consent (specific to stratum 2) from patients and/or their parents or legal guardian(s) obtained for study participation, data acquisition and transfer, collection and transfer of biological samples. Assent, of the child, when appropriate, will be obtained according to institutional guidelines, • Patients must be affiliated to a Social Security System in countries where this is mandatory.

4.3.2 Exclusion criteria • Tumour entity other than primary intracranial ependymoma, • Patients with WHO grade I ependymoma including myxopapillary variant , • Patients with spinal cord location of the primary tumour, • Participation within a different trial for treatment of ependymoma, • Concurrent treatment with any anti-tumour agents, • Inability to tolerate chemotherapy, • Unable to tolerate intravenous hydration, • Other severe acute or chronic medical or psychiatric conditions or laboratory abnormalities that may increase the risk associated with study participation or investigational product administration, or may interfere with the interpretation of study results in the judgment of the investigator, • Pre-existing mucositis, peptic ulcer, inflammatory bowel disease, ascites, or pleural effusion, • Contraindication to one of the IMP used in the stratum 2 according to the SmPCs in appendix 4 of this protocol ( SmPCs in appendix are those from UK which were chosen for the assessment of the safety aspects of the study ), • Patient for whom imaging remains RX despite all effort to clarify the MRI conclusion (see section Definitions of residual tumour ).

Page | 60 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

4.4 STRATUM 3

4.4.1 Inclusion criteria • Children younger than 12 months at time of entry to study or any patient ineligible to receive radiotherapy due to age at diagnosis, tumour location or clinician / parent decision and according to national criteria, • Newly diagnosed intracranial ependymoma of WHO grade II-III confirmed by central pathological review, • Adequate bone marrow function: o WBC > 2 x10 9/l o Peripheral absolute neutrophil count (ANC) ≥ 1 x 109/l o Platelet count ≥ 100 x 10 9/l, Hemoglobin > 8.0 g/dl (may have PRBC transfusion to achieve Hb > 8g/dl) • Adequate liver function: o Aspartate aminotransferase (AST) and/or alanine aminotransferase (ALT) ≤ 3 x institutional upper limit of normal for age o Total bilirubin ≤ 2 x institutional upper limit of normal for age • Adequate renal function: < 1.5 x Normal serum creatinine as adjusted for age: Age Maximum serum creatinine (mg/dl) < 5 years 0.8 Age Maximum serum creatinine (µmol/l) < 5 years 46 In case of suspected compromised renal function, glomerular filtration should be measured by an isotope GFR method such as EDTA clearance or by creatinine clearance. GFR should be: o > 80 ml/min/1.73m² for entry for children 2 years and over, o > 70 ml/min/1.73 m² for children aged 18-23 months, o > 65 ml/min/1.73 m² for children aged 12-17 months, o > 55 ml/min/1.73 m² for children of 6-11 months of age inclusive, o > 40 ml/min/1.73 m² for children of 0-5 months of age inclusive. • No previous chemotherapy (except steroids), • No previous radiotherapy, • No co-existent unrelated disease (e.g. renal, hematological) at the time of study entry that would render the patient unable to receive chemotherapy, • No medical contraindication to chemotherapy, • No signs of infection, • Patients and/or their parents or legal guardians must be willing and able to comply with scheduled visits, treatment plan, laboratory tests and other study procedures of the stratum 3, • Written informed consent (specific to stratum 3) from parents or legal guardian(s) obtained for study participation, data acquisition and transfer, collection and transfer of biological samples. Assent, of the child, when appropriate, will be obtained according to institutional guidelines, • Patients must be affiliated to a Social Security System in countries where this is mandatory.

Page | 61 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

4.4.2 Exclusion criteria • Tumour entity other than primary intracranial ependymoma, • Patients with WHO grade I ependymoma including myxopapillary variant , • Patients with spinal cord location of the primary tumour, • Participation within a different trial for treatment of ependymoma, • Concurrent treatment with any anti-tumour agents, • Inability to tolerate chemotherapy, • Unable to tolerate intravenous hydration, • Other severe acute or chronic medical or psychiatric conditions or laboratory abnormalities that may increase the risk associated with study participation or investigational product administration, or may interfere with the interpretation of study results in the judgment of the investigator, • Pre-existing mucositis, peptic ulcer, inflammatory bowel disease, ascites, or pleural effusion, • Pre-existing severe hepatic and/or renal damage, • Family history of severe epilepsy in immediate family siblings, • Presence of previously undiagnosed mitochondrial disorder detected by screening as part of trial, • Elevated blood ammonium level ≥ 1.5 x upper limit of the normal , • Elevated blood lactate level ≥ 1.5 x upper limit of the normal, • Contraindication to one of the IMP used in the stratum 3 according to the SmPCs in appendix 4 of this protocol (SmPCs in appendix are those from UK which were chosen for the assessment of the safety aspects of the study ).

4.5 Observational study All patients included in observational study must have been previously included in the staging phase or in one of the interventional strata.

4.5.1 Inclusion criteria • Patients who are not eligible to any of the 3 interventional studies proposal or who refuse to comply with protocol specifications of the interventional studies proposed, Nota Bene : Patients for whom imaging remains RX despite all effort to clarify the MRI conclusion (see Definitions of residual tumour ) and could not be recruited in stratum 3 can be included in the observational study • Main residence in one of the participating countries, • Age below 22 years old at the diagnosis, • Histological diagnosis of intracranial or spinal, localized or metastatic, ependymoma according to local pathologist (all WHO grades) including : myxopapillary ependymoma, ependymoma (papillary, clear-cell, tanycytic), ependymoma RELA fusion-positive and anaplastic ependymoma, confirmed by central pathological review Nota Bene: previously unregistered relapsing patients will be eligible, provided that the primary diagnosis is posterior to the study activation (Apr 29 th 2015), • All patients and/or their parents or legal guardians must receive an information sheet. Assent of the child, when appropriate, as well as written consent of parents or legal guardians will be obtained according to institutional / national guidelines.

Page | 62 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

4.5.2 Exclusion criteria • Tumour entity other than primary ependymoma • Patient with subependymomas and ependymoblastomas. • Primary diagnosis predating the activation of the SIOP Ependymoma II program (Apr 29 th 2015)

5 Consent

5.1 Screening

Patient who are potentially eligible for the study will be informed by the investigator about the different study requirements and the balance of risks and benefits of each study. If the patient or their parents/legal guardians agree to participate, they will be asked to give written informed consent. Patients will then be screened to ensure that they fulfil all selection criteria. There will be two steps of informed consent signatures: at staging and then, at the entry into the interventional strata. Patients participating in the observational study will receive a patient information sheet.

5.2 Informed Consent

Patients and / or parents or legal guardians will provide an appropriate written informed consent for each of the following parts of the trial: - the staging phase, - the subsequent interventional stratum - the integrated BIOMECA study. It is the responsibility of the investigator to obtain written informed consent for each patient prior to any trial related procedure. Patient/Parent Information Sheets (PIS) are provided prior to randomisation. Investigators must ensure that they adequately explain to the patient and/or parent the aim, trial treatment, anticipated benefits and potential hazards of taking part in the trial. It must be clearly explained to the patient/parents or legal guardian that participation in the biomarker studies, valproate pharmacokinetics and valproate pharmacodynamics is optional. The blood volumes will comply with EMA guidance for research in children. The time points for the biomarker studies will not require additional visits to hospital. No additional procedures under sedation/general anaesthesia will be performed. The Investigator should also stress that the patient is completely free to refuse to take part and if they do decide to take part they can withdraw from the trial at any time. The patient/parent or legal guardian should be given ample time to read each information sheet and to discuss their participation with others outside of the site research team. The patient/parent must be given an opportunity to ask questions which should be answered to their satisfaction. The right of the patient or parent to refuse to participate in the trial without giving a reason must be respected. If the patient/parent or legal guardian expresses an interest in their child participating in the trial they should be asked to sign and date the latest approved version of the informed consent form. The investigator must then sign and date the form on the same day as the patient and/or parent/legal guardian. According to local regulatory guidelines, written assent will also be obtained from the patient wherever it is possible. This assent form will also be signed and dated by the

Page | 63 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

investigator or their delegate. For those children who are unable to read, write or understand the assent form and where it is applicable, the clinician will explain the study and obtain verbal assent which will be documented in the medical notes. A copy of both the informed consent form (and assent forms when applicable) should be given to the patient/parent, a copy should be filed in the hospital notes, and the original placed in the Investigator Site File (ISF). Sufficient copies should be made to comply with local regulatory guidelines. Details of the informed consent discussions should be recorded in the patient’s medical notes, this should include date of, and information regarding, the initial discussion, the date consent was given, with the name of the trial and the version number of the patient information sheets and informed consent forms. Throughout the trial the patient/parent should have the opportunity to ask questions about the trial and any new information that may be relevant to the patient’s continued participation should be shared with them in a timely manner. On occasion it may be necessary to re- consent the patient or parent, for example, if new information becomes available or an amendment is made to the protocol that might affect the patient‘s participation in the trial. In this case the process above should be followed and the patient’s right to withdraw from the trial respected. Details of all preselected patients should be recorded on the Patient Screening/Enrolment Log. According to national practices and after the patient’s consent, their general practitioner should also be informed that they are taking part in the trial.

6 Trial Entry and Randomisation

6.1 Trial procedures Following brain and spinal MRI scan, patients will undergo surgery. As soon as the diagnosis of ependymoma is confirmed by local pathologist (before day 14), patients will be consented to take part in the SIOP II Ependymoma Program. All patients with intracranial tumour will undergo intracranial MRI. If not realised intra-operatively, post-operative MRIs will be performed within 72 hours following surgery (preferentially on day 1 or 2 after surgery). Spinal MRI will be performed post operatively if not done before surgery. Then, CSF collection will be obtained by lumbar puncture in all patients to evaluate metastasis status at day 14 and for research purposes if patient consented to take part to BIOMECA study. The national reference radiologist will proceed to central review of pre and postoperative MRI and operative reports. In parallel, the national reference pathologist will proceed to central pathological review to confirm the diagnosis and the grade of the tumour. Both national reference radiologist and pathologist will provide their reports to: • The investigational centre (PI) • The national investigator coordinator of the country concerned • The national coordinating centre of the country concerned If complete removal is confirmed in patients ≥ 12 months, the national coordinator will advise for inclusion in the phase III trial (stratum 1) and inform the local investigator. If residual disease is confirmed, the national coordinator will discuss the case with the local investigator (PI) concerned (and any others relevant expert: local radiologist, reference radiologist, local neuro surgeons, radiation oncologist) to decide on the feasibility of further surgery. If needed, the national coordinator will be responsible for referring the case to the national surgery referee or referees panel who will advise whether in their opinion the patient might be suitable for further surgery.

Page | 64 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

• If surgery is judged possible the national investigator and local investigator will discuss the benefit from second look surgery in his centre or, if preferred, in a centre with additional expertise in ependymoma surgery. Advice from national surgery referee or referees panel may be requested. • If further surgery is not possible, the national investigator coordinator will advise in the phase II study (stratum 2). The national investigator coordinator will advise for inclusion into the phase II trial and inform local investigator. After second surgery and according to its outcome, the eligibility to one of the 3 interventional strata would be re-evaluated as described above. For patients < 12 months who are not eligible to receive radiotherapy, a specific study to evaluate the efficacy of a dose intense chemotherapy to defer radiotherapy will be proposed (stratum 3); the national investigator coordinator will advise for inclusion into the phase II trial and inform local investigator.

6.2 Online randomisation procedures

Randomisation should be performed by on-sites investigator or designee using the online remote data entry system at the protocol specified time point. For each stratum, patient will be randomised in a 1:1 ratio. The stratum 1 randomisation will be stratified by tumour location (supra vs infratentorial), grading (classic vs anaplastic) and age (< vs ≥ 3 years at diagnosis).

Page | 65 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7 TREATMENT DETAILS

All of these IMPs described below are approved medicinal products which are commercially available within the EU. The use of all these IMPs but valproate is current standard of care for paediatric patients with ependymoma (See Trial Rationales ) All IMPS are held as standard hospital stock and should therefore be stored and handled according to local hospital policy. Chemotherapy drugs will not be provided by the Sponsor. Each treatment centre will be responsible for in house provision. The hospital pharmacists must maintain complete records of all study drugs. The labeling of all IMPs must be observed according to GCP and the applicable laws and regulations in the states concerned Institutional guidelines for the safe handling of anticancer drugs should be followed. Institutional guidelines for the safe handling of anticancer drugs should be followed. The investigator, the hospital pharmacist or other personnel allowed to store and supply investigational study drugs will be responsible for ensuring that the drugs are securely maintained as specified by the sponsor. Each participating country may use any IMP trade name authorised locally in the respect of the active substance and conditions of use described in in this protocol. Any strength is permitted. However, the formulations and the doses described in the protocol should be respected. All the following IMPs will be administered as infusion except valproate which should be taken as oral solution. The nature of the IMPs tested in the SIOP Ependymoma II program could evolve in the long run with the coming scientific progress and discoveries. Any changes will result in a new substantial amendment of the SIOP Ependymoma II program.

7.1 Stratum 1: Patients with no measurable residual disease and ≥ 12 months of age (Phase III)

7.1.1 Conformal radiotherapy This section of the protocol defines the guidelines for radiotherapy of the primary tumour site and must be utilized for all patients enrolled in the study. Radiotherapy must be delivered using conformal treatment planning and delivery techniques that fulfill the following : 1. 3-D imaging data (CT and/or MRI) acquisition with the patient in the treatment position. 2. Imaging data are utilized to delineate and reconstruct in three dimensions the gross tumour volume (GTV), the clinical target volume (CTV), the planning target volume (PTV), critical structures, and patient anatomy. 3. Beam geometry in three dimensions is defined on an individual basis taking into account structures traversed by each single beam as visualized by the beam’s eye view. 4. Dose distribution to the target and organs at risk is calculated on a point-by-point basis in the three dimensions, using a voxel-based tissue heterogeneity correction.

Page | 66 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.1.1.1 Timing of radiotherapy The interval between surgery and radiotherapy should be less than 6 weeks. Patients referred for adjuvant treatment after 6 weeks from surgery can be enrolled in the study. In this case a new MRI performed within 2 weeks prior radiotherapy start must confirm R0, R1 or R2 status.

7.1.1.2 Equipment • Megavoltage photons with a nominal energy ≥ 6MV must be used, • Intensity- modulated radiation therapy (IMRT) techniques are recommended, • Proton beams may be used, • Treatment with 60 Co is not permitted.

7.1.1.3 Target volumes Target volumes are defined according to the ICRU 50/62/78/83 Reports. MRI obtained immediately before radiotherapy should be used for treatment planning. To properly delineate target volumes for this study, complete information defining the extent of disease before and after surgery is needed. Pre- and post-operative MRI, in particular pre- and post-gadolinium contrast T1, T2, and FLAIR sequences, should be reviewed. Sequences that best define post- operative tumour bed and residual disease at each time point should be utilized to define the GTV and registered to planning CT.

Photon definition for GTV1, CTV1, and PTV1

• Gross Tumour Volume 1 (GTV1): the GTV includes the tumour bed at the primary site and macroscopic residual tumour after surgery. The pre-operative imaging defines all the tissues and anatomical areas initially involved with disease. The post-operative and pre-irradiation MRIs define the residual disease and the possibly collapsed post-surgical tumour bed which is the edge of the resection cavity. It is strongly suggested to register the best pre-operative MRI sequence showing tumour extension and to contour a structure denominated GTV1 pre-op to better identify relations between tumour and surrounding normal tissue anatomically involved. Tissue defects from surgical procedures should not be included in GTV1, if not involved by tumour. In case of discrepancies between intra-operative finding and imaging, the larger volume will define the GTV1.

• Clinical Target Volume 1 (CTV1): the CTV1 includes the GTV1 with an added margin to treat subclinical microscopic disease and is anatomically confined (i.e. the CTV1 is limited to the confines of the bony calvarium, falx and tentorium or extend up to but not beyond neuroanatomic structures through which tumour invasion is certain not to have occurred, including brainstem, and not invaded ventricle wall).). The CTV1 margin will be 0.5 cm for all patients but if the target volumes interface with the brainstem, the expansion of the CTV into the brainstem can be limited to 0,2 – 0,3 cm.

• Planning Target Volume 1 (PTV1): the PTV1 is a geometric expansion of the CTV1 to take into account for uncertainties in immobilisation, daily patient positioning and image registration. The PTV1 margin will be 0.2 cm-0.5 cm in all directions. The size of the required margins will depend on the quality of the immobilization device chosen and the accuracy and frequency of the imaging verification procedures available and adopted in the department for the patients enrolled in this protocol, and on the fulfillment of spinal cord constraints described in section 7.1.1.7.

Page | 67 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Photon definition for GTV2, CTV2, and PTV2 These target volumes are delineated with the purpose to administer the therapeutic dose to the volume containing the maximum concentration of tumor cells, limiting the dose to spinal cord, brainstem, and optic chiasm after the initial delivery of 54 Gy. GTV2 will be equal to GTV1 in case the sum plan dose distribution to spinal cord and optic chiasm will not exceed the recommendations in section 7.1.1.6. o Gross Tumor Volume2 (GTV2): 1. Spinal cord dose exceeded: GTV2 will be equal to the GTV1 minus the portion of the GTV1 caudal to the intracranial side of the foramen magnum 2. Optic chiasm dose exceeded: GTV2 will be equal to the GTV1 minus the optic chiasm with a planning organ at risk margin (thus generating a PRV) equal to that chosen for PTV1. 3. Spinal cord and optic chiasm dose exceeded: procedures 1 and 2 will be adopted o Clinical Target Volume2 (CTV2): the CTV2 includes the GTV2 with an added margin of 0.5 cm and is anatomically confined. The margin will be reduced to 0.2-0.3 cm in correspondence of the brain stem and to 0.0 cm in correspondence of the spinal cord and/or the optic chiasm PRV when dose to these critical structures must be limited to 54 Gy. o Planning Target Volume2 (PTV2): will include the CTV2 with an additional margin equal to the margin used to create the PTV1 or equal to 0.0 cm in correspondence of the spinal cord and/or the optic chiasm PRV when dose to these critical structures must be limited to 54 Gy. PTV2 will be equal to PTV1 in case the sum plan dose distribution to spinal cord and optic chiasm will not exceed the recommendations in section 7.1.1.6

Proton definition for GTV1, CTV1, and PTV1 o Gross Tumour Volume 1 (GTV1): is the same for photons and protons. o Clinical Target Volume 1 (CTV1): is the same for photons and protons. o Planning Target Volume 1 (PTV1): for protons both lateral margins and the margin in depth (relative to proximal and distal tumour surface) have to be considered. Consequently for each beam orientation the PTV with appropriate margins laterally and along the beam direction will take into account range uncertainties and tissue in homogeneities. Any individual adaptation is at the discretion of the local team (radiotherapist and physicist).

Proton definition for GTV2, CTV2, and PTV2 o Gross Tumor Volume2 (GTV2): 1. Spinal cord dose exceeded: GTV2 will be equal to the GTV1 minus the portion of the GTV1 caudal to the intracranial side of the foramen magnum 2. Optic chiasm dose exceeded: GTV2 will be equal to the GTV1 minus the optic chiasm with a planning organ at risk margin (thus generating a PRV) equal to that chosen for PTV1. 3. Spinal cord and optic chiasm dose exceeded: procedures 1 and 2 will be adopted

o Clinical Target Volume2 (CTV2): the CTV2 includes the GTV2 with an added margin of 0.5 cm and is anatomically confined. The margin will be reduced to 0.2-0.3 cm in correspondence of the brain stem and to 0.0 cm in correspondence of the spinal cord and/or the optic chiasm PRV when dose to these critical structures must be limited to 54 Gy.

Page | 68 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

o Planning Target Volume2 (PTV2): will include the CTV2 with an additional margin equal to the margin used to create the PTV1 or equal to 0.0 cm in correspondence of the spinal cord and/or the optic chiasm PRV when dose to these critical structures must be limited to 54 Gy. PTV2 will be equal to PTV1 in case the sum plan dose distribution to spinal cord and optic chiasm will not exceed the recommendations in section 7.1.1.6

Organs at risk All possible organs at risk must be identified and contoured, in particular: • Infratentorial brain • Brain stem • Hypophysis • Hypothalamus • Cochlea (bilateral) • Lens (bilateral) • Optic chiasm • Optic nerves (bilateral) • Supratentorial brain • Temporal lobes (bilateral) • Hippocampus (bilateral) • Spinal cord • Thyroid

7.1.1.4 Dosimetry Photon dose is specified in dose-to-water (Gy). Proton dose is prescribed in cobalt Gray equivalent (Gy 1.1) using a recommended value of radiobiological effectiveness (RBE) of 1.1 [Paganetti 2002]. For simplicity all proton doses referred to in this protocol are the RBE-weighted proton absorbed dose DRBE and are the dose of photons that would produce the same therapeutic effect as a proton absorbed-dose, given under identical circumstances [ICRU 2007]. Prescription point: for photon beams the prescription point is at or near the isocenter unless an IMRT technique is adopted. Total dose: the total dose to the PTV1 is 54 Gy in 30 daily fractions of 1,8 Gy, and total dose to PTV2 is 5,4 Gy in 3 daily fractions of 1,8 Gy. In case GTV1 is equal to GTV2 (PTV1 = PTV2), total dose to the PTV1 will be 59,4 Gy in 33 daily fractions of 1,8 Gy. Children < 18 months (and > 12 months) at irradiation without post-surgical residual disease or children with risk factors, namely multiple surgeries (more than 2) or poor neurological status will receive 54 Gy. Fractionation: all patients will receive a daily fraction of 1.8 Gy, five fractions per week. Dose-Volume prescription: at least 95% of the protocol specified dose should encompass 98% of the PTV1/2. Dose-Volume uniformity: no more than 10% of the PTV1/2 should receive 110% of the protocol dose as evaluated by DVH.

Page | 69 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Attention should be paid to avoid 110% isodose involving normal tissue outside the PTV1/2. 110% isodose distribution should be confined to the PTV1/2.

7.1.1.5 Treatment technique Patient position: patients may be treated in the supine or prone position. Use of immobilization devices is mandatory. General anesthesia is strongly recommended for younger children. Treatment planning images: a treatment planning CT with the patient in the treatment position is required. CT slices thickness should be ≤ 4 mm, preferably 2 mm, and the images should be obtained as close as possible to the start of radiotherapy. Contrast is not mandatory but strongly recommended. MR registration: registration of MR to treatment planning CT is mandatory for all patients. Treatment planning: computerized treatment planning systems for 3D conformal radiotherapy must be utilized (DVH, BEV, DRR). Beam geometries and treatment techniques must be selected to minimize dose to critical organs without compromising dose homogeneity to PTV1/2. Proton beams: for posterior fossa treatment, rotating gantries seem to be advisable as compared to horizontal beam lines being only capable of achieving lateral beam arrangements (non-coplanar, vertex fields are acceptable, but rotating beams technique are preferred). As there is some uncertainty about increased RBE at the distal Bragg peak, weighting of spots and Bragg peaks need to be carefully evaluated. The use of multiple field techniques might be preferred in order to avoid cumulating high weighted spots in critical areas. As with conventional treatment, organ tolerances as well as target coverage are to be respected. Beam shaping: All photon beams must be individually shaped with shielding at least 5HVL thick or multi-leaf collimation. If proton beams are delivered using passive scattering technology: appropriate beam energy, range of spread out Bragg peak, collimators and compensators are individualized for each beam. If spot scanning technology is used, conformality will be achieved by treatment planning, software steering and control systems. Hardware accessories will not necessarily need to be used. However, attention should be paid to achieve sharp penumbra and appropriate energy (reducing of air gap, slim range shifter, avoiding neutron production etc.). Treatment verification: regular treatment verification according to institutional policies is required. As a minimum standard, weekly portal images must be performed and the set-up variations recorded. Rest: There will be no planned rest or breaks of treatment.

7.1.1.6 Dose to Organs at Risk • Spinal cord: the spinal cord must be contoured from the inferior border of the foramen magnum to a length of at least 6 cm caudally. For infratentorial tumours extending beyond the foramen magnum, the corresponding spinal cord will be excluded at a cumulative physical dose of 54 Gy. In all other cases, the cervical spinal cord possibly included in the PTV will be excluded at a cumulative physical dose of 50 Gy.

• Brain Stem: the brain stem must be contoured from the superior border of the foramen magnum to the upper aspect of the mesencephalon (inferior to the third ventricle and optic

Page | 70 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

tract). During treatment planning procedures, any effort should be made to spare as much as possible at least part of the brainstem from the highest dose level.

Dose constraints for the spinal cord and the brainstem, adopted by the SFOP Radiotherapy Quality Assurance Working Group, are reported in the following 2 Tables. They refer to two different clinical Ependymoma presentations with the tumor developing inside the posterior fossa or below the foramen magnum, respectively.

Page | 71 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

• Children with infratentorial tumours undergoing more than 2 surgeries or neurologically compromised are already planned to receive 54 Gy.

• Optic chiasm: should be contoured on at least 2 successive planning CT slices. Optic chiasm will be excluded at a cumulative treatment dose of 54 Gy.

• Cochlea: each cochlea will be contoured on at least 2 successive slices of planning CT as a circular structure within the petrous portion of the temporal bone. The mean dose to at least one cochlea should be limited to 30 Gy.

7.1.1.7 Quality Control of Radiotherapy (QC) General organisation • Radiotherapy QC will be organised and undertaken on a national basis. This should include a procedure to reproduce and check target volumes and dose prescription. • Each national group will appoint a national radiotherapy coordinator. • The national radiotherapy QC coordinator will work in close co-operation with two to three named radiotherapy colleagues forming the national radiotherapy QC panel. This will ensure the constant availability of a QC assessor without delay. • Submission of planning documentation – The use of DICOM RT compatible data entailing the treatment plan, conventional imaging or computer generated dose distributions may vary from country to country. Each national group will decide on the most appropriate way of submitting films and plans (i.e. either electronically or via courier). Guidance can be obtained from the national radiotherapy coordinator.

Page | 72 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

• Countries who do not wish to set up their own QC panel should at the start of the study identify the QC panel of a national group of their choice, which - if agreed - will provide the prospective QC for them.

Documentation to provide Copies of the following data should be sent to the national radiotherapy QC panel (preferably by electronic transfer of digitised images): • All the diagnostic imaging used to plan the target volume, including MRI prior to surgery of the primary tumour. • Radiotherapy record including prescription and daily and cumulative doses to target and OAR. • Digitally reconstructed radiographs (DRR) or simulator film for each treatment field and orthogonal (anterior-posterior and lateral) images for isocenter localization for each group of concurrently treated beam. When using IMRT, only orthogonal isocenter images are required. • Isodose distribution for the composite treatment plan in axial, sagittal and coronal planes at the center of the treatment or planning target volume. • Dose volume histograms (DVH) for the composite treatment plan for all target volumes and required OAR. Treatment planning system summary report that includes the monitor units calculation, beam parameters, calculation algorithm, and volume of interest dose statistics. • Beams-eye-view (BEV) of portals showing collimator, beam aperture, target volume and critical structures are required when not using IMRT.

7.1.1.8 Definition of protocol deviation When delivering radiotherapy to the tumour bed. Prescription dose: Minor deviation: difference in prescribed or computed dose is > 5 and ≤ 10% of protocol specified dose. Major deviation: difference in prescribed or computed dose is > 10% of protocol specified dose. Dose Uniformity: Minor deviation: 95% isodose to ≥ 95% and < 100% of PTV

110% to > 10% but ≤ 15% of PTV Major deviation: 95% isodose to < 95% of PTV 110 % to > 15% of PTV Volume: Minor deviation: CTV and PTV margins are less than those specified in the protocol in the absence of anatomic barriers to tumour invasion Major deviation: GTV does not encompass MRI visible residual tumour.

7.1.1.9 Treatment modifications due to hematological Toxicity According to institutional guidelines (optional), full blood count should be taken in all patients once a week during radiotherapy. Treatment will not be interrupted for anaemia, leucopoenia or thrombocytopenia unless life threatening as per CTCAE v4.03 . Blood product or growth factor support should be instituted according to institutional guidelines.

Page | 73 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.1.2 Maintenance chemotherapy The following drugs are IMPs in this stratum:

IMP Formulation Vincristine Infusion or injection according to national guidelines Etoposide Infusion Cyclophosphamide Infusion Cisplatin Infusion

Patients will be randomised to receive or not 16 weeks of VEC+cisplatin. Chemotherapy with VEC plus cisplatin must be started 4 weeks after the completion of radiotherapy and MRI assessment.

VEC+ cisplatin course D1: Vincristine (VCR): 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus D1-D3: Etoposide (VP16): 100 mg/m² infused over 1 – 4 hours according Week 1 to standard institutional practice. Days 1 - 3 Cycle 1 D1: Cyclophosphamide: 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. Week 4 CISPLATIN (CDDP): 80 mg/m² as an IV infusion over 4 hours. Day 22 VINCRISTINE (VCR): 1.5 mg/m² (maximal dose 2 mg) i.v. bolus D36: Vincristine (VCR): 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus D36-D38 : Etoposide (VP16): 100 mg/m² infused over 1 – 4 hours Week 6 according to standard institutional practice. Days 36 - 38 Cycle 2 D36: Cyclophosphamide: 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. Week 9 CISPLATIN (CDDP) 80 mg/ m² as an IV infusion over 4 hours ... Day 57 VINCRISTINE (VCR) : 1.5 mg/m² (maximal dose 2 mg) i.v. bolus D71: Vincristine (VCR) : 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus D71-D73 Etoposide (VP16) : 100 mg/m² infused over 1 – 4 hours Week 11 according to standard institutional practice. Days 71 - 73 Cycle 3 D71 Cyclophosphamide: 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. Week 14 CISPLATIN (CDDP) 80 mg/ m² as an IV infusion over 4 hours … Day 92 VINCRISTINE (VCR): 1.5 mg/ m² (maximal dose 2 mg) i.v. bolus D106: Vincristine (VCR): 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus Week 16 D106-D108 Etoposide (VP16 ): 100 mg/m² infused over 1 – 4 hours Cycle 4 Days 106 - according to standard institutional practice. 108 D106 Cyclophosphamide: 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals.

Table 3: Maintenance chemotherapy schedule for patients enrolled in the Stratum 1

The body surface area must be capped at 2.00 m² for any calculation of the IMP dose to be administered.

Page | 74 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

For babies less than 10 kg, doses will be based on body weight (BW) rather than body surface area (BSA). The following doses will be used: • VINCRISTINE (VCR): 0.05 mg/kg as an i.v. bolus • ETOPOSIDE (VP16): 3.3 mg/kg infused over 1 – 4 hours according to standard institutional practice. • CYCLOPHOSPHAMIDE (CPM): 100mg/kg in 3 divided infusions (33mg/kg/infusion) infused over 60 minutes at eight hourly intervals. • CISPLATIN (CDDP) 2.7 mg/kg. as an IV infusion over 4 hours.

7.1.2.1 Vincristine Vincristine should always be given according to national guidelines with respect to the avoidance of inadvertent administration of vinca alkaloids by other than the IV route.

7.1.2.2 Etoposide Etoposide may be administered according to standard institutional practice with respect to the vehicle for infusion and the infusion time (within the specified limits of 1 – 4 hours). To avoid fluid overload, the total daily hydration volume should be reduced by the volume in which the etoposide is administered.

7.1.2.3 Cyclophosphamide To prevent haemorrhagic cystitis, cyclophosphamide should be administered with mesna (sodium 2- mercaptoethane sulfonate) and hydration according to standard institutional practice. The total daily dose (total 3000mg/m²) is split into three doses of 1000mg/m² administered at T=1, 9 & 17 hours (every 8 hours). Cyclophosphamide may be prepared in accordance with standard institutional practice with respect to the vehicle for infusion and concentration, or as follows: • < 1000 mg in 50 ml. • 1000 mg to 2000 mg in 100 ml.

• Pre-hydration Pre-hydration should be administered at 125 ml/m²/hour using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium chloride 20mmol/L, for 3 hours before the first cyclophosphamide infusion.

• MESNA Administration of MESNA should start at the same time as pre-hydration and, once started, run continuously until 12 hours after completion of final (third) cyclophosphamide infusion. The total dose of MESNA should be based on 120% of the total dose of cyclophosphamide (3600mg/m²) and be infused over 32 hours. It may be administered as a separate infusion or added to the pre- hydration and hydration fluid according to standard institutional practice.

• Hydration Administration of hydration should start at the same time as cyclophosphamide and run at 125 ml/m²/hour (3000ml/m²/24 hours) using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium chloride 20mmol/L. Once started, it should run continuously until at least 12 hours after completion of final (third) cyclophosphamide infusion (total of at least 29 hours) .

Page | 75 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.1.2.4 Cisplatin Pre and post hydration may be given according to standard institutional practice or:

T=0 hour: Pre-hydration 2.5 % glucose/0.45 % sodium chloride should be administered at a rate of 200 ml/m²/hour for 3 hours. Because it is critical to maintain a good urine output prior to cisplatin administration, urine output should be monitored and should not be allowed to fall below 3 ml/kg/h for 2 hours. Should output fall below this level, a bolus of mannitol (0.5 g/kg over 15 to 30min) with additional fluid bolus of 10 ml/kg (2.5 % dextrose /0.45 % sodium chloride) should be administered. Furosemide should not be administered since it may impair renal clearance of cisplatin.

T=3 hours: Cisplatin infusion over 4 hours : Cisplatin may be prepared in accordance with standard institutional practice with respect to the vehicle for infusion and concentration, or as follows: • 10–50 mg in 100 mL • 50–100 mg in 150 mL • >100 mg in 250mL

T=3 hours: Hydration during and until 8 hours post cisplatin (i.e. infuse over 12 hours) Administration of hydration should start at the same time as cisplatin and run at 125 ml/m²/hour (1500 ml/m²/12 hours) using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with mannitol 12 g/L. and potassium chloride 20mmol/L. Once started, it should run continuously for at least 12 hours.

T=15 hours: Subsequent hydration until 24 hours post cisplatin Administration of post-hydration should start 12 hours after the start of the cisplatin infusion and run at 125 ml/m²/hour (2000 ml/m²/16 hours, maximum volume 3375 ml) using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium 20mmol/L, magnesium 10mmol/L. and calcium 0.6mmol/L. Once started, it should run continuously for at least 16 hours.

7.1.2.5 Investigations during chemotherapy • Full Blood Count should be checked weekly. Before administration of chemotherapy neutrophils must be > 1 x 10 9/L and platelets > 100 x 10 9/L. • Serum creatinine should be checked before starting treatment and before each course of chemotherapy. − If the serum creatinine is higher than the normal for age then glomerular filtration rate should be formally measured by 51 Cr EDTA clearance before administering cisplatin. − If isotopic clearance techniques are not available creatinine clearance should be measured. − Treatment should be delayed until renal function recovers if the corrected GFR is: ° For patient over the age of two years GFR should be > 80 ml/min/1.73 m². ° For children of 18-23 months of age inclusive GFR should be > 70 ml/min/1.73 m².

Page | 76 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

° For children of 12-17 months of age inclusive GFR should be > 65 ml/min/1.73 m² • Serum sodium, potassium, calcium, magnesium, bilirubin, AST and ALT must be measured before each administration of chemotherapy. • Potential neurological damages due to chemotherapy will be followed up during study participation (please refer to section 10.6 Audiogram or BAER ).

7.2 Stratum 2: Patients with inoperable measurable residue and ≥ 12 months of age (Phase II)

7.2.1 Definition or Investigational Medicinal Products (IMPs) The following drugs are IMPs in this stratum:

IMP Formulation Vincristine Infusion or injection according to national guidelines Etoposide Infusion Cyclophosphamide Infusion Cisplatin Infusion Methotrexate Infusion

7.2.2 VEC+ High dose Methotrexate arm Induction chemotherapy with VEC intensified with HD-MTX must be started as early as possible after routine surgical recovery (ideally within 3 weeks of surgery). Patients will receive HD-MTX at a dose of 8000 mg/m² as an IV infusion over 24 hours on Days 15, 36 and 57. 10% of the dose should be given over the first hour and the remaining 90% over 23 hours, immediately following the first infusion of 10%. The second infusion must finish at T=25 (24 hours after the first infusion was started) even if it has not been completed at that time. After 3 cycles of VEC+HD-MTX, the patient will be assessed for response and a central review of imaging will assess the possibility for the patient to undergo second look surgery if this is deemed possible .

VEC+HD-MTX course

D1: Vincristine (VCR) : 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus Week 1 D1-D3 Etoposide (VP16) : 100 mg/m² infused over 1 – 4 hours according to standard institutional practice. Days 1 - 3 D1 Cyclophosphamide : 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. Cycle 1 Methotrexate: 8000mg/m² as an IV infusion over 24 hours Week 3 10% of the dose should be given over the first hour and the remaining 90% over the following 23 hours. The second infusion must finish at Day 15 T=25 (24 hours after the first infusion was started) even if it has not been completed at that time.

Page | 77 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

D22: Vincristine (VCR) : 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus Week 4 D22-D24: Etoposide (VP16) : 100 mg/m² infused over 1 – 4 hours according to standard institutional practice. Days 22 - 24 D22: Cyclophosphamide : 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. Cycle 2 Methotrexate: 8000mg/m² as an IV infusion over 24 hours. Week 6 10% of the dose should be given over the first hour and the remaining 90% over the following 23 hours. The second infusion must finish at Day 36 T=25 (24 hours after the first infusion was started) even if it has not been completed at that time. D43: Vincristine (VCR) : 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus Week 7 D43-D45: Etoposide (VP16) : 100 mg/m² infused over 1 – 4 hours according to standard institutional practice.. Days 43 - 45 D43: Cyclophosphamide : 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. Cycle 3 Methotrexate: 8000mg/m² as an IV infusion over 24 hours.

Week 9 10% of the dose should be given over the first hour and the remaining 90% over the following 23 hours. The second infusion must finish at Day 57 T=25 (24 hours after the first infusion was started) even if it has not been completed at that time. Table 4: VEC+HD-MTX schedule for patients enrolled in the Stratum 2

The body surface area must be capped at 2.00 m² for any calculation of the IMP dose to be administered. For babies less than 10 Kg, doses will be based on body weight (BW) rather than body surface area (BSA). The following doses should be used: • VINCRISTINE (VCR): 0.05 mg/kg as an i.v. bolus • ETOPOSIDE (VP16): 3.3 mg/kg infused over 1 – 4 hours according to standard institutional practice. • CYCLOPHOSPHAMIDE (CPM): 100 mg/kg in 3 divided infusions i.e. 33 mg/kg/ infusion infused over 60 minutes at eight hourly intervals. • There is no dose adjustment for the doses of IV methotrexate.

7.2.3 VEC arm Frontline postoperative CT with VEC must be started as early as possible after routine surgical recovery (ideally within 3 weeks of surgery). 3 cycles of VEC will be administered every 3 weeks.

VEC course

D1: Vincristine (VCR): 1.5mg/m² (Maximum dose: 2mg) as an IV bolus Cycle 1 D1-D3: Etoposide (VP16): 100mg/m² infused over 1 – 4 hours according Week 1 to standard institutional practice. Days 1 - 3 D1: Cyclophosphamide: 3000mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. D22: Vincristine (VCR): 1.5mg/m² (Maximum dose: 2mg) as an IV bolus Cycle 2 D22-D24: Etoposide (VP16): 100mg/m² infused over 1 – 4 hours

Page | 78 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Week 4 according to standard institutional practice . Days 22 - 24 D22: Cyclophosphamide: 3000mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. D43: Vincristine (VCR): 1.5mg/m² (Maximum dose: 2mg) as an IV bolus Cycle 3 D43-D45: Etoposide (VP16): 100mg/m² infused over 1 – 4 hours Week 7 according to standard institutional practice. Days 43 - 45 D43: Cyclophosphamide: 3000mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. Table 5: VEC schedule for patients enrolled in the Stratum 2

The body surface area must be capped at 2.00 m² for any calculation of the IMP dose to be administered.

For babies weighing less than 10 kg, doses will be based on body weight (BW) rather than body surface area (BSA). The following doses should be used: • VINCRISTINE (VCR): 0.05 mg/kg as an i.v. bolus • ETOPOSIDE (VP16): 3.3 mg/kg infused over 1 – 4 hours according to standard institutional practice. • CYCLOPHOSPHAMIDE (CPM): 100mg/kg in 3 divided infusions (33mg/kg/infusion) infused over 60 minutes at eight hourly intervals .

7.2.3.1 Recommended guidelines for the administration of drugs

7.2.3.1.1 Vincristine Vincristine should always be given according to national guidelines with respect to the avoidance of inadvertent administration of vinca alkaloids by other than the IV route.

7.2.3.1.2 Etoposide Etoposide may be administered according to standard institutional practice with respect to the vehicle for infusion and the infusion time (within the specified limits of 1 – 4 hours). To avoid fluid overload, the total daily hydration volume should be reduced by the volume in which the etoposide is administered

7.2.3.1.3 Cyclophosphamide To prevent the risk of hemorrhagic cystitis, cyclophosphamide should be administered with mesna (sodium 2-mercaptoethane sulfonate) and hydration according to standard institutional practice. The total daily dose (total 3000mg/m²) is split into three doses of 1000mg/m² administered at T=1, 9 & 17 hours (every 8 hours). Cyclophosphamide may be prepared in accordance with standard institutional practice with respect to the vehicle for infusion and concentration, or as follows: − < 1000 mg in 50 ml. − 1000 mg to 2000 mg in 100 ml.

• Prehydration Pre-hydration should be administered at 125 ml/m²/hour using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium chloride 20mmol/L, for 3 hours before the first cyclophosphamide infusion.

Page | 79 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

• MESNA Administration of MESNA should start at the same time as pre-hydration and, once started, run continuously until 12 hours after completion of final (third) cyclophosphamide infusion. The total dose of MESNA should be based on 120% of the total dose of cyclophosphamide (3600mg/m²) and be infused over 32 hours. It may be administered as a separate infusion or added to the pre- hydration and hydration fluid according to standard institutional practice.

• Hydration Administration of hydration should start at the same time as cyclophosphamide and run at 125 ml/m²/hour (3000ml/m²/24 hours) using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium chloride 20mmol/L. Once started, it should run continuously until at least 12 hours after completion of final (third) cyclophosphamide infusion (total of at least 29 hours).

7.2.3.1.4 Methotrexate Methotrexate may be prepared in accordance with standard institutional practice with respect to the vehicle for infusion and concentration, or as follows: • <2400 mg in 100 ml, • 2400 mg to 6000 mg in 250 ml, • >6000 mg in 500 ml

• Prehydration Pre-hydration should be administered at 200ml/m²/hour using an infusion fluid according to standard institutional practice, or 4% glucose/0.18% sodium chloride with sodium bicarbonate 50mmol/L. and potassium chloride 20mmol/L. Sodium bicarbonate may be administered as a separate infusion if this is in accordance with standard institutional practice. Pre-hydration should be administered for at least 3 hours before the first methotrexate infusion (10%). The first methotrexate infusion (10%) should only start when a urinary pH of >7.0 is established.

• Hydration Administration of hydration should start at the same time as the first dose of methotrexate (10%) and run at 125 ml/m²/hour (3000ml/m²/24 hours) using an infusion fluid according to standard institutional practice, or 4% glucose/0.18% sodium chloride with sodium bicarbonate 50mmol/L. and potassium chloride 20mmol/L. Sodium bicarbonate may be administered as a separate infusion if this is in accordance with standard institutional practice. Once started, hydration should run continuously until MTX level is below 0.1 µmol /l.

• Schedule for Folinic Acid Rescue : Folinic acid 15 mg/m² as an IV bolus. Folinic acid rescue should commence 36 hours after the start of the first methotrexate (10%) infusion. Initially, five doses should be administered every three hours. Thereafter, folinic acid should be given every six hours until the MTX level < 0.1 µmol/L. (< 1 X 10 -7 molar). The MTX level should be measured at 48, 72 and 96 hours after the start of the first infusion (10%), and every 24 hours thereafter until the MTX level is < 0.1 µmol/L. (< 1 X 10 -7 molar).

Page | 80 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Folinic acid rescue should continue until methotrexate level is < 0.1 µmol / l (< 1 X 10 -7 molar) in order to ensure that there is adequate neuroprotection and prevention of skin toxicity with the high cumulative dose of methotrexate recommended in this protocol. A minimum number of doses is not specified.

Time after Methotrexate plasma concentration ( µmol /l) starting methotrexate < 0.1 0.1-2 2-20 20-100 >100 48h None a 15 mg/ m² 6 h 15 mg/ m² 6 h 10 mg/ m² 3 h 100 mg/ m² 3 h 72h None 15 mg/ m² 6 h 10 mg/ m² 3 h 100 mg/ m² 3 h 1 g/m 2 3h 96h None 15 mg/m 2 6 h 10 mg/m 2 3 h 100 mg/m 2 3 h 1 g/m 2 3h 120h b None 15 mg/m 2 6 h 10 mg/m 2 3 h 100 mg/m 2 3 h 1 g/m 2 3h

a) No extra folinic acid is required provided MTX levels are below 0.1 µmol /l (10 -7 M) at 48 hours b) At time points after 120h folinic acid administration should be continued as recommended for 120h. Table 6: folinic rescue schedule In the event of renal toxicity associated with HD-MTX and consequent delayed excretion of MTX, folinic acid rescue should continue as specified in the above table. If serum creatinine rises by > 100% of baseline within 24 hours of MTX administration, additional supportive care should be discussed as follows: − Hemofiltration or haemodialysis should be considered if renal function continues to deteriorate and should be administered according to the advice of the local experts in intensive care and nephrology. It should be noted that the plasma level of MTX may appear to fall quickly after haemodialysis and will also fall after haemofiltration. However the tissue level will remain high and continued folinic acid rescue is necessary until haemodialysis or haemofiltration stops and the steady state level of MTX can be measured. − Glucarpidase (if available for use within participating countries) may be considered, as an alternate to hemofiltration or haemodialysis, in a situation of deteriorating renal function and prolonged excretion of MTX. Glucarpidase is a recombinant glutamate carboxypeptidase that hydrolyses MTX to inactive metabolites. It should be given as a single dose of 50 units/kg as a slow intravenous injection over 3-5 minutes. Folinic acid should be stopped 4 hours before glucarpidase and then reintroduced 4 hours after glucarpidase administration at the dose and frequency indicated by the pre-glucarpidase plasma methotrexate level. Plasma MTX levels should decrease by >98% in 15 minutes but, as with haemodialysis and haemofiltration, tissue levels will initially remain high. Elimination takes 8 hours. The DAMPA metabolite of MTX is known to cross-react with methotrexate in immunoassay analysis of MTX levels resulting in an overestimation of plasma methotrexate levels after glucarpidase. For this reason HPLC assays should be used if available. Plasma levels of MTX should be continually monitored they fall below 0.1 µmol/l. Hypersensitivity reactions to glucarpidase are rare with the first dose. Repeat administration carries a high risk of allergy or lack of efficacy due to tolerance.

If it has been necessary to use haemofiltration, haemodialysis or glucarpidase, no further administration of high-dose methotrexate is permitted in this protocol.

Page | 81 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Figure 1: Appropriate Folinic Acid Rescue according to Plasma Methotrexate Level

1000 Plasma MTX 2 1000mg/m 2 every 3 hours 100mg/m level for plateau every 3 hours of line on graph

100 100 micromol 10mg/m 2 every 3 100mg/m 2 every 3 hours hours

20 micromol 15mg/m 2 every 6 hours 10mg/m 2 every 3 hours 10

15mg/m 2 every 6 hours 2 micromol 1 Plasma MTX(micromoles/L) Plasma

0,1 0.1 micromol Expected Rate of Fall. No further folinic acid rescue once level < 0.1 micromol /l

0,01 36 48 60 72 84 96 108 120 132 144 Time since Start of methotrexate administration (Hou rs)

Page | 82 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.2.3.2 Investigations during chemotherapy • Full Blood Count (FBC) should be checked weekly: Before administration of VEC chemotherapy, neutrophils must be > 1 x 10 9/l and platelets > 100 x 10 9/l. • Cycles of HD-MTX may be given on schedule even if there is neutropenia or thrombocytopenia provided that the patient is well and free from infection. Serum levels of MTX should be monitored as described above • Serum creatinine should be checked before starting treatment and before each course of chemotherapy. o If the serum creatinine is higher than the normal for age or if there is delayed elimination of methotrexate then glomerular filtration rate should be formally measured by 51 Cr EDTA clearance. o If isotopic clearance techniques are not available creatinine clearance should be measured. o Treatment should be delayed until renal function recovers if the corrected GFR is: ° For patient over the age of two years GFR should be > 80 ml/min/1.73 m 2. ° For children of 18-23 months of age inclusive GFR should be > 70 ml/min/1.73 m 2. ° For children of 12-17 months of age inclusive GFR should be > 65 ml/min/1.73 m². • Serum sodium, potassium, calcium, magnesium, bilirubin, AST and ALT must be measured before each administration of chemotherapy. o Provided serum bilirubin is within normal limits MTX can be given if serum AST and ALT are elevated to not more than 10 X the upper limit of normal. • Potential neurological damages due to chemotherapy will be followed during study participation (please refer to section 10.6 Audiogram or BAER ).

7.2.4 Conformal radiotherapy All patients will receive then a conformal radiotherapy after chemotherapy (please refer to section 7.1.1 ).

7.2.5 Radiotherapy boost A radiotherapy boost of 8 Gy will be administered to the residual measurable tumour immediately after the completion of the conformal radiotherapy in the following cases: • for patients with residual measurable disease that remains unresectable despite frontline chemotherapy and • for whom second line surgery is not feasible, and • with remaining residual measurable tumour despite standard conformal radiotherapy.

All patients with measurable residual disease after irradiation of the tumour bed that is detected by neuroradiological studies are eligible for this section of the study. A residue is considered to be any area of pathological signal and/or enhancement comparable with the appearance of the preoperative tumour and measurable in 3 planes on MRI. Definition of residual disease and MRI sequences required to delineate the target of radiotherapy boost are detailed in section Radiological assessments . For patients eligible to this section of the stratum, central neuroradiological, neurosurgical, and radiotherapy reviews are mandatory to verify radiation boost indications and feasibility.

Page | 83 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Patients without evidence of residual disease at neuroradiological studies after the window study and/or a second-look surgery and conformal radiotherapy are not eligible for radiotherapy boost.

7.2.5.1 Requirements to deliver the boost ° Conformal radiation therapy treatment planning and delivery techniques applicable to small target volumes ensuring a highly homogeneous dose distribution with a dose variation within the PTV between +/- 5% of the prescription dose ° Image registration tools ° Reliable devices for patient immobilization ° Cone-beam CT mandatory daily For each potential centre willing to participate to the SIOP Ependymoma II program, a feasibility assessment regarding their ability to perform boost of radiotherapy will be realized by the national coordinating centre before any site initiation visit. For each non-qualified centre, the nearest qualified centre meeting the protocol requirements could be designated by the national coordinating centre to perform whole radiotherapy including boost in the respect of the time window imposed by the protocol (see below).

Imaging required for radiotherapy boost treatment planning: during the last week of conformal radiotherapy to the tumour bed or soon after the radiotherapy end, a new diagnostic MRI and a planning CT with contrast must be obtained for all patients and the two studies must be registered for GTV delineation.

• Timing of radiotherapy boost: the radiotherapy boost must be delivered soon after the end of the conformal radiotherapy phase. A delay of a maximum of 7 days can be acceptable if required to organize the new diagnostic MRI and planning CT with contrast: this could be the case especially for the younger children requiring general anesthesia.

• Target Volumes: Gross Tumour Volume (GTV): the GTV includes all the measurable pathological tissue with or without contrast enhancement after surgery/ies as documented by MRI at the end of conformal treatment to tumour bed.

• Clinical target Volume (CTV): CTV is equal to the GTV, i.e. no additional margin to be applied to GTV for CTV. This is to restrict the boost volume as much as possible, in view of the likely adjacency of the brainstem for most patients, the hypofractionated schedule and the resulting cumulative biological effective dose

• Planning Target Volume (PTV): the PTV is a geometric expansion of the GTV/CTV to take into account for uncertainty in immobilisation, daily patient positioning and image registration. The PTV margin is 0.1-0.3 cm in all directions according to local policies.

• Organ at Risk: the brain stem must be contoured from the superior border of the foramen magnum to the upper aspect of the mesencencephalon as detected on the most cranial planning CT slice. No more than 1.5 cm² of brain stem (around 5% of total brain stem volume) should be included in the PTV boost and therefore no more than 1.5 cm² of brain stem should receive the prescribed boost dose. This criterion is derived from the clinical experience gathered within the Italian AIEOP Ependymoma Protocol, recently closed and published (Massimino et al. 2016).

Page | 84 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

All possible organs at risk must be identified and contoured as listed in section 7.1.1.3 Target volumes.

• Dosimetry: the total dose is 8 Gy. Two fractions of 4 Gy will be administered in two consecutive treatment days. Dose variations in the PTV must be within -5% - +5% of the prescription dose. The entire PTV should be encompassed within the 95% isodose surface.

• Treatment technique: a stereotactic, conformal (image guided, multiple field) PT and IMRT techniques are recommended.

• Patient position: patients must be treated in the supine position. Use of immobilization devices is mandatory. Deep sedation or general anesthesia is strongly recommended for younger children.

• Treatment planning images: a treatment planning CT (or MRI) with the patient in the treatment position is required. CT slices thickness must be ≤ 2 mm and the study must be obtained during the last week or the immediately subsequent one of conformal radiotherapy. Planning CT images must be registered with diagnostic MRI images obtained in the same time point (last or subsequent week of conformal radiotherapy phase).

7.2.5.2 Quality Control of Radiotherapy (QC) General Organisation and Documentation to provide: please refer to section 7.1.1.7 .

7.2.5.3 Definition of protocol deviation Ù Radiotherapy boost to post-surgical residual disease Dose Uniformity Minor deviation: 95% isodose to ≥ 95% and < 100% of PTV 105% to > 5% but ≤ 10% of PTV Major deviation: 95% isodose to < 95% of PTV Or 105% to >10% of PTV Volume Minor deviation: PTV margins are less than those specified in the protocol. Major deviation: GTV does not encompass MRI visible residual tumour.

7.2.5.4 Treatment Modifications due to hematological Toxicity According to institutional guidelines (optional): in all patients full blood count should be taken once a week during radiotherapy. Treatment will not be interrupted for anaemia, leucopoenia or thrombocytopenia unless life threatening. Blood product or growth factor support should be instituted according to institutional guidelines.

Page | 85 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.2.6 Maintenance chemotherapy Provided patients did not show progression with the frontline schedule, patients will receive 16 weeks of VEC–CDDP. Maintenance chemotherapy must be started 4 weeks after the completion of radiotherapy.

VEC+ cisplatin course D1: Vincristine (VCR): 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus D1-D3: Etoposide (VP16): 100 mg/m² infused over 1 – 4 hours according Week 1 to standard institutional practice. Days 1 - 3 Cycle 1 D1: Cyclophosphamide: 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. Week 4 CISPLATIN (CDDP): 80 mg/m² as an IV infusion over 4 hours. Day 22 VINCRISTINE (VCR): 1.5 mg/m² (maximal dose 2 mg) i.v. bolus D36: Vincristine (VCR): 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus D36-D38 : Etoposide (VP16): 100 mg/m² infused over 1 – 4 hours Week 6 according to standard institutional practice. Days 36 - 38 Cycle 2 D36: Cyclophosphamide: 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. Week 9 CISPLATIN (CDDP) 80 mg/ m² as an IV infusion over 4 hours ... Day 57 VINCRISTINE (VCR) : 1.5 mg/m² (maximal dose 2 mg) i.v. bolus D71: Vincristine (VCR) : 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus D71-D73 Etoposide (VP16) : 100 mg/m² infused over 1 – 4 hours Week 11 according to standard institutional practice. Days 71 - 73 Cycle 3 D71 Cyclophosphamide: 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals. Week 14 CISPLATIN (CDDP) 80 mg/ m² as an IV infusion over 4 hours … Day 92 VINCRISTINE (VCR): 1.5 mg/ m² (maximal dose 2 mg) i.v. bolus D106: Vincristine (VCR): 1.5 mg/m² (maximal dose 2 mg) as an i.v. bolus Week 16 D106-D108 Etoposide (VP16 ): 100 mg/m² infused over 1 – 4 hours Cycle 4 Days 106 - according to standard institutional practice. 108 D106 Cyclophosphamide: 3000 mg/m² in 3 divided infusions (1000 mg/m²/infusion) infused over 60 minutes at eight hourly intervals.

Table 7: Maintenance chemotherapy schedule for patients enrolled in the Stratum 2

For babies weighing less than 10 kg, doses will be based on body weight (BW) rather than body surface area (BSA). The following doses should be used: • VINCRISTINE (VCR): 0.05 mg/kg as an i.v. bolus • ETOPOSIDE (VP16): 3.3 mg/kg over 1 – 4 hours according to standard institutional practice. • CYCLOPHOSPHAMIDE (CPM): 100mg/kg in 3 divided infusion i.e. 33 mg/kg/ infusion given over 60 minutes at eight hourly intervals. • CISPLATIN (CDDP) 2.7 mg/kg as an IV infusion over 4 hours.

7.2.6.1 Vincristine Vincristine should always be given according to national guidelines with respect to the avoidance of inadvertent administration of vinca alkaloids by other than the IV route.

Page | 86 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.2.6.2 Etoposide Etoposide may be administered according to standard institutional practice with respect to the vehicle for infusion and the infusion time (within the specified limits of 1 – 4 hours). To avoid fluid overload, the total daily hydration volume should be reduced by the volume in which the etoposide is administered.

7.2.6.3 Cyclophosphamide To prevent haemorrhagic cystitis cyclophosphamide should be administered with MESNA (sodium 2- mercaptoethane sulfonate) and hydration according to standard institutional practice. The total daily dose (total 3000mg/m²) is split into three doses of 1000mg/m² administered at T=1, 9 & 17 hours (every 8 hours). Cyclophosphamide may be prepared in accordance with standard institutional practice with respect to the vehicle for infusion and concentration, or as follows: • < 1000 mg in 50 ml. • 1000 mg to 2000 mg in 100 ml.

• Pre-hydration Pre-hydration should be administered at 125 ml/m²/hour using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium chloride 20mmol/L, for 3 hours before the first cyclophosphamide infusion.

• MESNA Administration of MESNA should start at the same time as pre-hydration and, once started, run continuously until 12 hours after completion of final (third) cyclophosphamide infusion. The total dose of MESNA should be based on 120% of the total dose of cyclophosphamide (3600mg/m²) and be infused over 32 hours. It may be administered as a separate infusion or added to the pre- hydration and hydration fluid according to standard institutional practice.

• Hydration Administration of hydration should start at the same time as cyclophosphamide and run at 125 ml/m²/hour (3000ml/m²/24 hours) using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium chloride 20mmol/L. Once started, it should run continuously until at least 12 hours after completion of final (third) cyclophosphamide infusion (total of at least 29 hours) .

7.2.6.4 Cisplatin Pre- and post-hydration may be given according to standard institutional practice or:

T= 0 hour: Prehydration 2.5 % glucose/0.45 % sodium chloride should be administered at a rate of 200 ml/m²/hour for 3 hours. Because it is critical to maintain a good urine output prior to cisplatin administration, urine output should be monitored and should not be allowed to fall below 3ml/kg/hour for 2 hours. Should output fall below this level a bolus of mannitol (0.5g/kg over 15 to 30min.) with an additional fluid bolus of 10ml/kg should be administered. Furosemide should NOT be administered since it may impair renal clearance of cisplatin.

Page | 87 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

T= 3 hours: Cisplatin infusion over 4 hours: Cisplatin may be prepared in accordance with standard institutional practice with respect to the vehicle for infusion and concentration, or as follows: ° 10 – 50 mg in 100 mL ° 50 – 100 mg in 150 mL ° >100mg in 250mL T= 3 hours: Hydration during and until 8 hours post cisplatin (i.e. infuse over 12 hours) Administration of hydration should start at the same time as cisplatin and run at 125 ml/m²/hour (1500 ml/m²/12 hours) using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with mannitol 12 g/L. and potassium chloride 20mmol/L. Once started, it should run continuously for at least 12 hours. T= 15 hours: Subsequent post hydration until 24 hours post cisplatin Administration of post-hydration should start 12 hours after the start of the cisplatin infusion and run at 125 ml/m²/hour (2000 ml/m²/16 hours, maximum volume 3375 ml/m²) using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium 20mmol/L, magnesium 10mmol/L. and calcium 0.6mmol/L. Once started, it should run continuously for at least 16 hours.

7.2.6.5 Investigations during chemotherapy • Full Blood Count should be checked weekly. Before administration of chemotherapy neutrophils must be > 1 x 10 9/L and platelets > 100 x 10 9/L. • Serum creatinine should be checked before starting treatment and before each course of chemotherapy. − If the serum creatinine is higher than the normal for age then glomerular filtration rate should be formally measured by 51 Cr EDTA clearance before administering cisplatin. − If isotopic clearance techniques are not available creatinine clearance should be measured. − Treatment should be delayed until renal function recovers if the corrected GFR is: ° For patient over the age of two years GFR should be > 80 ml/min/1.73 m 2. ° For children of 18-23 months of age inclusive GFR should be > 70 ml/min/1.73 m 2. ° For children of 12-17 months of age inclusive GFR should be > 65 ml/min/1.73 m2. • Serum sodium, potassium, calcium, magnesium, bilirubin, AST and ALT must be measured before each administration of chemotherapy. • Potential neurological damage due to chemotherapy will be followed during study participation (please refer to section 10.6 Audiogram or BAER ).

Page | 88 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.3 Stratum 3: Randomized Phase II chemotherapy study in children < 12 months or those not eligible to receive radiotherapy

7.3.1 Definition or Investigational Medicinal Products (IMPs) The following drugs are IMPs in this stratum:

IMP Formulation (Sodium) Valproate Oral solution Methotrexate Infusion Vincristine Infusion or injection according to national guidelines Cyclophosphamide Infusion Carboplatin Infusion Cisplatin Infusion

7.3.2 Post-operative intensive Chemotherapy Stratum 3 is open to all children with intracranial ependymoma WHO grade II and III who are ineligible to receive radiotherapy due to age at diagnosis, tumour location or clinician / parent decision and according to national criteria. Patients do not need to have a complete resection prior to entry in the study, delayed or second look surgery is an acceptable part of the strategy and chemotherapy induced responses should be carefully documented. Assessment of response to chemotherapy in patients with a tumour residuum will be collected as part of trial data. Patients will be randomised to receive a dose intense chemotherapy alternating myelosuppressive and relatively non-myelosuppressive drugs at 2 weekly intervals with or without the addition of Sodium valproate as a Histone Deacetylase inhibitor. Chemotherapy is administered in pulses at 14 days intervals between Days 1 and 380. Patients with residual disease after the completion of treatment, progressive disease during treatment or relapse during or after treatment should be considered for surgery and / or radiotherapy. All patients need to be screened prior to the commencement of valproate to exclude the possibility of undiagnosed very rare mitochondrial disorders: personal and direct familial history of severe epilepsy in other siblings and persisting elevated blood lactate and ammonemia will be evaluated prior to inclusion and 3 months after starting therapy. Sodium Valproate is a relatively weak Histone deacetylase inhibitor (HDACi), it is therefore important to aim to achieve relatively high serum levels [100-150 µg/ml]. Based on phase II data and the British National Formulary, we chose a twice daily administration. However, early indications suggest that some children need to have three time daily dosing schedule to achieve therapeutic levels of VPA when used as an HDACi . We suggest you start on a twice a day (bisdiua/ b.d) dosing schema moving to three times a day dosing schedule (t.d.s) administration if therapeutic levels of VPA are not being achieved [100-150µg/ml]. For example, if therapeutic levels are not achieved when giving 60 mg/kg/day (30 mg/kg b.d) then the total daily VPA dose is divided and given 3 times daily that is 20mg/kg t.d.s. If therapeutic levels are still not achieved increase to higher total daily doses, in 5 mg/kg/day increments and again administer in three divided doses (that is 65 mg /kg/day then 70 mg /kg/day etc). We recommend an interval of a week between each change in dose level giving time to recheck therapeutic levels on new dose level. If a dose is

Page | 89 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

increased and the patient experiences toxicity then when VPA is recommenced this should be at the previously tolerated dose level.

CYCLE N° 1 2 3 4 5 6 7 Vincristine and Carboplatin D1 D 57 D113 D169 D225 D281 D337 Vincristine and Methotrexate D15 D 71 D127 D183 D239 D295 D351 Vincristine and Cyclophosphamide D29 D 85 D141 D197 D253 D309 D365 Cisplatin 2-day continuous infusion D43-4 D99-100 D154-5 D211-2 D267-8 D323-4 D379-80

Valproate Initial dose: 30 mg /kg/day for two weeks in 2 divided doses (15mg/kg / b.d.) (for the experimental arm only) Increasing weekly up to 40->50->60 mg /kg/day in 2 divided doses. If therapeutic levels 100-150µg/ml are not achieved when giving 60 mg /kg/day administer in three divided doses (20mg/kg/dose b.d.). If therapeutic level is still not achieved increase to higher total daily dose in 5 mg/kg/day increments that is 65 mg /kg/day (21.6mg/kg/dose t.d.s) then 70 mg /kg/day (23.3mg/kg/dose t.d.s) in 3 divided doses. Reassessment ** ** **

**If patient has residual disease, they should be considered for further surgery at each reassessment point. Table 8a: Post-operative intensive Chemotherapy schedule for patients enrolled in the Stratum 3

Dosing Schedule (**) Dose for infants over Dose for infants 6 to Dose for infants less than 6 12 months 11 months months

Vincristine 1.5 mg/m² x 1 1.125 mg/m² x 1 0.75 mg/m² x 1 Carboplatin 550 mg/m²x 1 412.5 mg/m² x 1 275 mg/m² x 1

Vincristine 1.5 mg/m² x 1 1.125 mg/m² x 1 0.75 mg/m² x 1 Methotrexate 8000 mg/m² x 1 6000 mg/m² x 1 4000 mg/m² x 1 Folinic Acid [See below. No dose reduction]

Vincristine 1.5 mg/m² x 1 1.125 mg/m² x 1 0.75 mg/m² x 1 [maximum dose 2mg] Cyclophosphamide 1500 mg/m² x 1 1125 mg/m² x 1 750 mg/m² x 1 Mesna [See below]

Cisplatin 80mg/m²/48 hours 60mg/m²/48 hours 40mg/m²/48 hours

Valproate 30 mg/kg/day* 30 mg/kg/day* 30 mg/kg/day* *initial dosing then, according to monitoring, increasing up to 60 mg/kg/day in 2 divided doses (30mg/kg/dose b.d.) [see 7.3.2.6] ** for patients aged: • 12 months and over : full surface area based dose of chemotherapy • 6-11 months : 75% of the surface-area-based dose of chemotherapy • Under 6 months : 50% of the surface-area –based dose of chemotherapy IF THERAPEUTIC LEVELS 100-150 µg/ml ARE NOT ACHIEVED: when giving 60 mg/kg/day we recommend that VPA is administered in three divided doses for a further week (20mg/kg/dose t.d.s) then trough levels checked – if levels remain sub-therapeutic increase to next dose level i.e.65mg/kg/day in three divided doses. Table 8b: Dosing schedule according to the age of patients enrolled in the Stratum 3

Page | 90 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.3.2.1 Vincristine Vincristine should always be given according to national guidelines with respect to the avoidance of inadvertent administration of vinca alkaloids by other than the IV route.

7.3.2.2 Carboplatin Carboplatin should be administered as an IV infusion over 1 hour. It may be prepared in accordance with standard institutional practice with respect to the vehicle for infusion and concentration, or as follows: ° 25 to 50 mg in 50 ml ° 50 to 500 mg in 100 ml

7.3.2.3 Methotrexate Methotrexate is given as an IV infusion over a total of 24 hours. 10 % of dose is given over 1 hour while the remaining 90 % of the dose should be given over the following 23 hours. For doses administered, please refer to table above. The infusion must finish at 24 hours even if it has not been completed. Methotrexate may be prepared in accordance with standard institutional practice with respect to the vehicle for infusion and concentration, or as follows: • < 2400mg in 100ml • 2400mg to 6000mg in 250ml • > 6000mg in 500ml

• Pre-hydration Pre-hydration should be administered at 200ml/m²/hour using an infusion fluid according to standard institutional practice, or 4% glucose/0.18% sodium chloride with sodium bicarbonate 50mmol/L. and potassium chloride 20mmol/L. Sodium bicarbonate may be administered as a separate infusion if this is in accordance with standard institutional practice. Pre-hydration should be administered for 3 - 6 hours before the first methotrexate infusion (10%). The first methotrexate infusion may start after 3 hours of pre-hydration provided that a urinary pH of >7.0 is established.

• Hydration Administration of hydration should start at the same time as the first dose of methotrexate (10%) and run at 125 ml/m²/hour (3000ml/m²/24 hours) using an infusion fluid according to standard institutional practice, or 4% glucose/0.18% sodium chloride with sodium bicarbonate 50mmol/L. and potassium chloride 20mmol/L. Sodium bicarbonate may be administered as a separate infusion if this is in accordance with standard institutional practice. The concentration and/or rate of sodium bicarbonate administration should be adjusted to maintain a urinary pH of > 7.0 Once started, hydration should run continuously until methotrexate level is below 0.1 µmol/l.

• Schedule for Folinic Acid Rescue Folinic acid: 15mg/m² as an IV bolus. Folinic acid rescue should commence 36 hours after the start of the first methotrexate (10%) infusion. Initially, five doses should be administered every three hours. Thereafter, folinic acid should be given every six hours until the MTX level < 0.1 µmol/L. (< 1 X 10 -7 molar).

Page | 91 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

The MTX level should be measured at 48, 72 and 96 hours after the start of the first infusion (10%), and every 24 hours thereafter until the MTX level is < 0.1 µmol/L. (< 1 X 10 -7 molar). Folinic acid rescue should continue until the MTX level is < 0.1 µmol/L. (< 1 X 10 -7 molar) in order to ensure that there is adequate neuroprotection and prevention of skin toxicity with the high cumulative dose of methotrexate recommended in this protocol. A minimum number of doses is not specified.

Time after Methotrexate plasma concentration (µM /L) starting methotrexate < 0.1 0.1-2 2-20 20-100 >100 48h None a 15 mg/m² 6 h 15 mg/m² 6 h 10 mg/m² 3 h 100 mg/m² 3 h 72h None 15 mg/m²6 h 10 mg/m² 3 h 100 mg/m² 3 h 1 g/m² 3h 96h None 15 mg/m² 6 h 10 mg/m² 3 h 100 mg/m² 3 h 1 g/m² 3h 120h b None 15 mg/m² 6 h 10 mg/m² 3 h 100 mg/m² 3 h 1 g/m² 3h (a) No extra folinic acid is required provided MTX levels are below 0.1 µmol /l (10 -7 µM) at 48 hours (b) At time points after 120h folinic acid administration should be continued as recommended for 120h

Table 9: folinic rescue schedule

See also Figure 1: appropriate folinic acid rescue according to plasma Methotrexate level

In the event of renal toxicity associated with HD-MTX and consequent delayed excretion of MTX, folinic acid rescue should continue as specified in the above table. If serum creatinine rises by > 100% of baseline within 24 hours of MTX administration, additional supportive care should be discussed as follows:

- Haemofiltration or haemodialysis should be considered if renal function continues to deteriorate and should be administered according to the advice of the local experts in intensive care and nephrology. It should be noted that the plasma level of MTX may appear to fall quickly after haemodialysis and will also fall after haemofiltration. However the tissue level will remain high and continued folinic acid rescue is necessary until haemodialysis or haemofiltration stops and the steady state level of MTX can be measured.

- Glucarpidase (if available for use within participating countries) may be considered, as an alternate to haemofiltration or haemodialysis, in a situation of deteriorating renal function and prolonged excretion of MTX. Glucarpidase is a recombinant glutamate carboxypeptidase that hydrolyses MTX to inactive metabolites. It should be given as a single dose of 50 units/kg as a slow intravenous injection over 3-5 minutes. Folinic acid should be stopped 4 hours before glucarpidase and then reintroduced 4 hours after glucarpidase administration at the dose and frequency indicated by the pre-glucarpidase plasma methotrexate level. Plasma MTX levels should decrease by >98% in 15 minutes but, as with haemodialysis and haemofiltration, tissue levels will initially remain high. Elimination takes 8 hours. The DAMPA metabolite of MTX is known to cross-react with methotrexate in immunoassay analysis of MTX levels resulting in an overestimation of plasma methotrexate levels after glucarpidase. For this reason HPLC assays should be used if available. Plasma levels of MTX should be continually monitored they fall below 0.1 µmol/l. Hypersensitivity reactions to glucarpidase are rare with the first dose. Repeat administration carries a high risk of allergy or lack of efficacy due to tolerance.

If it has been necessary to use haemofiltration, haemodialysis or glucarpidase, no further administration of high-dose methotrexate is permitted in this protocol.

Page | 92 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.3.2.4 Cyclophosphamide To prevent haemorrhagic cystitis cyclophosphamide should be administered with MESNA (sodium 2- mercaptoethane sulfonate) and hydration according to standard institutional practice. Cyclophosphamide is given as an IV infusion over 1 hour. Cyclophosphamide may be prepared in accordance with standard institutional practice with respect to the vehicle for infusion and concentration, or as follows:

• < 1000mg in 50ml • 1000mg to 2000mg in 100ml • Pre-hydration Pre-hydration should be administered at 125 ml/m²/hour using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium chloride 20mmol/L, for 3 hours before the first cyclophosphamide infusion.

• MESNA Administration of MESNA should start at the same time as pre-hydration and, once started, run continuously until 12 hours after completion of the cyclophosphamide infusion. The total dose of MESNA should be based on 120% of the total dose of cyclophosphamide: 1800mg/m² (> 1 year.); 1350mg/m² (6-12 months) or 900mg/m² (< 6 months). It may be administered as a separate infusion or added to the pre- hydration and hydration fluid according to standard institutional practice.

• Hydration Administration of hydration should start at the same time as cyclophosphamide and run at 125 ml/m²/hour (3000ml/m²/24 hours) using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium chloride 20mmol/L. Once started, it should run continuously until at least 12 hours after completion of the cyclophosphamide infusion.

7.3.2.5 Cisplatin (Note this is different to the regime in VEC-Cisplatin) Cisplatin is given as an IV infusion over 48 hours. Pre- and post-hydration may be given according to standard institutional practice or: T=0 hour: Pre-hydration: 2.5 % glucose/0.45 % sodium chloride should be administered at a rate of 200 ml/m²/hour for 3 hours. Because it is critical to maintain a good urine output prior to cisplatin administration, urine output should be monitored and should not be allowed to fall below 3ml/kg/hour for 2 hours. Should output fall below this level a bolus of mannitol (0.5g/kg over 15 to 30 min) with an additional fluid bolus of 10ml/kg should be administered. Furosemide should NOT be administered since it may impair renal clearance of cisplatin. T=3 hours: Cisplatin infusion over 48 hours: Cisplatin may be prepared in accordance with standard institutional practice with respect to the vehicle for infusion and concentration, or as follows:

• 10 – 50mg in 100 ml • 50 – 100mg in 150 ml

Page | 93 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

• >100mg in 250ml

T=3 hours: Hydration during and until 6 hours post cisplatin (i.e. infuse over 54 hours) Administration of hydration should start at the same time as cisplatin and run at 125 ml/m²/hour (3000ml/m²/24 hours) using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with mannitol 12 g/L. and potassium chloride 20mmol/L. Once started, it should run continuously for 54 hours. T=57 hours: Post-hydration until at least 24 hours post cisplatin Administration of post-hydration should commence on completion of the preceding hydration and continue for 18 hours (24 hours after the end of the cisplatin infusion). It should run at 125 ml/m²/hour (2250ml/m²/18 hours) using an infusion fluid according to standard institutional practice, or 2.5% glucose/0.45% sodium chloride with potassium 20mmol/L, magnesium 10mmol/L. and calcium 0.6mmol/L. Once started, it should run continuously for at least 18 hours.

7.3.2.6 Valproate Patients randomised to receive valproate will be given valproate from the start of standard dose intense chemotherapy and during one year after the last chemotherapy administration as a maintenance treatment. In case of disease progression during chemotherapy or during maintenance VPA treatment, valproate must be discontinued. Valproate will be initially administered in a twice daily dosing schedule by mouth, using sodium valproate oral solution. To avoid somnolence, it is recommended that the starting dose should be 30mg/kg in two divided doses that is 15mg/kg twice a day. This may then be increased as follows in order to achieve target trough plasma concentrations of [100-150µg/ml]: • First weekly increment: 20 mg/kg twice daily (total daily dose = 40 mg/kg) • Second weekly increment: 25 mg/kg twice daily (total daily dose = 50 mg/kg) • Third weekly increment: 30 mg/kg twice daily (total daily dose = 60 mg/kg)

For example: A 15kg child starts treatment at 30mg/kg/day and receives 450mg a day in two divided doses that is 225mg twice a day. An increase in dose to 40mg/kg/day would give a dose of 300mg twice daily (600mg per day). Wait at least one week at each dose level before checking trough level and only increase the dose if the trough level < 100 µg/ml. Based on previous studies it is suggested that trough levels in the range 100 - 150μg/ml. (700μM to 1040μM) should be aimed for. At this target trough level there will be 20- 50 % free valproate. If patients tolerate 60mg/kg/day that is 30mg/kg given twice daily at the third weekly increment and trough serum levels are below 100-150 µg/ml, then we recommend changing to a three times a day dosing regimen so that the total daily VPA dose is divided and given three times daily that is 20mg/kg t.d.s. If therapeutic levels are still not achieved one week after this change, we recommend an increase to higher total daily doses in 5mg/kg/day increments and again administer in three divided doses that is 65mg/kg/day (21.6mg/kg/dose t.d.s). Again, if therapeutic levels are still not achieved one week after this change, increase the dose to 70mg/kg/day that is 23.3mg/kg/dose t.d.s. Continue in 5mg/kg/day dose intervals till therapeutic levels are achieved. We recommend an interval of a week between change in dose level giving time to recheck therapeutic levels on a new dose level.

Page | 94 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Careful assessment of Full Blood Count is very important as higher valproate doses may affect bone marrow function. For pharmacokinetics and pharmacodynamics studies, please refer to Appendix 6 – Valproate Pharmacokinetics and Pharmacodynamics studies (optional) .

Dose Limiting Toxicities related to valproate Dose limiting toxicities have been described in the Phase 1 study [77] at valproate trough levels of 100 to 150µg/ml: predominantly of somnolence with isolated cases of pulmonary hemorrhage, dyspnea, hypoxia, hypotension and hypocalcaemia. At the lower trough levels of 75 to 100µg/ml, no DLTs were reported. Ependymoma cell line studies and preclinical work (H. Witt Personal communication) suggest that there is a benefit to higher dose levels. In the Phase I study a partial response was seen in a thalamic GBM at 75 to 100µg/ml, and a minor response in a DIPG at 100-150 µg/ml [77]. Given that we are also delivering chemotherapy alongside valproate we may not be able to achieve the proposed higher trough level of 100-150 µg/ml. If patients initially tolerate valproate and are escalated to higher doses than 30mg/kg twice daily and unacceptable toxicity is observed then the dose of valproate should be reduced to the dose level that was previously tolerated by the patient and serum levels, full blood count and symptom review repeated at the end of that week. Alternatively try to give the same dose in three rather than two divided doses as above. Serum levels of Valproate need to be monitored every month to ensure patients are on appropriate dose level and this information will be recorded in the data set.

Table of Valproate dose modification guidelines Somnolence is a side effect of VPA, but is likely to be relatively uncommon in young children. It is important that we only reduce dose for objective reasons, a novel grading scheme has been introduced to help guide Valproic acid dosing: Grade I: mild Somnolence; increased sleepiness but not affecting not affecting normal activities Ù no change in medication Grade II: moderate Somnolence; Increased duration of sleep, more drowsy when asleep, less playful and active Ù If persistent for more than 1 week reduce dose of Sodium Valproate by 20% or back to previous dose level that had been tolerated previously

Grade III: marked somnolence: difficult to rouse from sleep, falling asleep when normally awake and active, unwilling to play. Ù Reduce dose of Sodium Valproate by 20% or back to previous dose level that had been tolerated. Ù If persist more than one week , discontinue valproate If a child develops somnolence: Serum amylase, Ammonia, Acylcarnitine, Lactate, Urine organic acids, EEG must be checked.

Page | 95 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

In case of side effect likely related to VPA (e.g. tremor, low platelets, weight gain, pancreatitis, etc.); please refer to the CTCAE v 4.03 for grading. VPA dose modification should follow the following recommendations: Grade I: no change in medication Grade II: If persistent for more than 1 week reduce dose of Sodium Valproate by 20% or back to previous dose level that had been tolerated previously Grade ≥ III: Reduce dose of Sodium Valproate by 20% or back to previous dose level that had been tolerated. If persist more than one week, discontinue valproate

Body weight Dose (mg) First Dose (mg) Second Dose Third Dose (kg) twice daily increment twice (mg) increment (mg) increment daily twice daily twice daily 4 60 80 100 120

5 75 100 125 150

6 90 120 150 180 7 105 140 175 210 8 120 160 200 240

9 135 180 225 270

10 150 200 250 300 11 165 220 275 330 12 180 240 300 360

13 195 260 325 390

14 210 280 350 420 15 225 300 375 450 16 240 320 400 480

17 255 340 425 510 18 270 360 450 540 19 285 380 475 570

20 300 400 500 600

Table 10: Valproate dosing using a twice daily regimen

Please NOTE If dose increases do not appear to be mirrored in an increase in VPA tough level to reach required therapeutic level when a dose of 60mg/kg/day has been reached then we recommend that the frequency of VPA medication is changed to three times daily at the same dose level, that is 60mg/kg/day or 20mg/kg per dose three times daily. If levels have not increased to a therapeutic level of 100-150 µg/ml by this point, increase to next dose level i.e. 65mg/kg/day (21.6mg/kg/dose) in three divided doses. Continue to increase dose by 5mg/kg/day till therapeutic level is achieved or toxicity is encountered.

Page | 96 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

7.3.2.7 Investigations during chemotherapy • Full Blood Count should be checked prior to each cycle or administration of chemotherapy. Before the carboplatin administration of each cycle neutrophils must be > 1 x 10 9/L and platelets > 100 x 10 9/L. • We would recommend that the cytokine G-CSF is used to maintain dose intensity; this may be used on a pro re nata (P.R.N) basis • Subsequent courses within that cycle can be given on schedule even if there is neutropenia or thrombocytopenia provided that the child is well and free from infection. • Serum creatinine should be checked before starting treatment and before each course of chemotherapy. o If the serum creatinine is higher than the normal for age or if there is delayed excretion of methotrexate then glomerular filtration rate should be formally measured by 51 Cr EDTA clearance before carboplatin, methotrexate or cisplatin. o If isotopic clearance techniques are not available creatinine clearance should be measured. o Treatment should be delayed until renal function recovers if the corrected GFR is below the levels shown. ° For children over the age of two years GFR should be > 80 ml/min/1.73 m². ° For children of 18-23 months of age inclusive GFR should be > 70 ml/min/1.73 m². ° For children of 12-17 months of age inclusive GFR should be > 65 ml/min/1.73 m². ° For children of 6-11 months of age inclusive GFR should be > 55 ml/min/1.73 m². ° For children of 0-5 months of age inclusive GFR should be > 40 ml/min/1.73 m². o It is recommended that glomerular filtration rate should be formally measured before the first, third, fifth, sixth and seventh cycles of treatment. • Serum sodium, potassium, calcium, magnesium, bilirubin, AST and/or ALT must be measured before each administration of chemotherapy. Provided serum bilirubin is within normal limits methotrexate can be given if serum AST and ALT are elevated to not more than 10 X the upper limit of normal. • Potential audiological deterioration due to chemotherapy will be followed during study participation (please refer to section 10.6 Audiogram or BAER).

7.3.2.8 Indication for radiotherapy Disease progression and viable residual disease proven by histology of residual disease at the end of chemotherapy are considered as indication for using radiation as salvage therapy. If there is no evidence of residual disease or of residual viable disease, then there is no indication for further therapy and close observation with 3 monthly MRI scans is part of this strategy. Radiotherapy in the absence of residual viable disease or of disease progression is a protocol violation.

Page | 97 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

8 MODIFICATION OF DOSE CHEMOTHERAPIES The following guidelines on the monitoring of toxicity of chemotherapy and modification of the dose of chemotherapy are recommendations. For additional advice, contact the national coordinator.

8.1 Hematology A full blood count should be performed every week after the start of each course of chemotherapy.

Before each course VEC: WBC < 2 x 10 9/L or Delay chemotherapy for at least one week Neutrophils < 1 x 10 9/L or Platelets < 100 x 10 9/L Platelet/WBC recovery delays therapy > 2 Omit one day of etoposide from future courses weeks and / or severe aplasia of VEC (i.e. infusion on Days 1 & 2 only) If delays more than 3 weeks Contact national coordinator (*)

Before each course vincristine + CDDP: WBC < 2 x 10 9/L or Delay chemotherapy for at least one week. Neutrophils < 1 x 10 9/L or Platelets < 100 x 10 9/L Platelet/WBC recovery delays therapy > 2 Omit the course of Vincristine + cisplatin weeks and / or severe aplasia If delays more than 3 weeks Contact National coordinator (*)

(*) Haematological toxicities requiring chemotherapy delays of more than 3 weeks should be discussed on a case by case with the national coordinator of the study in the country concerned. The national coordinator should ensure that the correct doses have been administered, that any concomitant viral disease is excluded and that the use of any concomitant haematotoxic compound is excluded as a cause. Any decision on the timetable and procedures to be followed should be considered once all the potential causes of the hematological toxicity have been considered. The national coordinator should ask for additional investigations if needed and report the case to chief investigators for clinical advice and final decisions.

8.2 Neurotoxicity of vincristine

Epileptic seizure or Ileus Stop vincristine in this course. Reduce vincristine to 1 mg/m² in the next course. After recovery Aim to give vincristine at 100% doses as tolerated Significant dysesthesia, muscle weakness Omit vincristine until recovery or abdominal pain After recovery Aim to give vincristine at 100% doses as tolerated.

Page | 98 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

8.3 Nephrotoxicity of Cisplatin:

• Nephrotoxicity is a major toxicity concern with cisplatin. Both glomerular and tubular toxicity must be monitored during treatment with cisplatin. • Dose modification is based on glomerular toxicity i.e. a reduction in glomerular filtration rate (GFR). • In the context of a multicentre international study, it is appreciated that different national groups and individual centres have varying but well-established methods of measuring or estimating GFR. These include methods based on blood clearance of radioisotope, e.g. MAG3 and clearance of 51 Cr EDTA or Tc99m DTPA, estimation of creatinine clearance from the plasma creatinine level (e.g. using the Schwartz formula) or by direct measurement of urinary creatinine clearance. • Any well-established method of a estimating GFR as detailed above may be used prior to initiating chemotherapy. However, if the serum creatinine is higher than normal for age before any subsequent courses of treatment, it is recommended that blood clearance of radioisotope, or direct measurement of urinary creatinine clearance should be performed before each cisplatin containing course.

8.3.1 Cisplatin dose modification used with VEC in Strata 1 and 2 The normal range for creatinine clearance should be considered to be: • 80 ml/min/1.73m² for children 2 years and over, • 70 ml/min/1.73 m² for children aged 18-23 months, • 65 ml/min/1.73 m² for children aged 12-17 months old.

The cisplatin dose modification guidelines are the followings:

Serum creatinine > 1.5 x ULN or GFR / Creatinine clearance: • Age 2 years and over < 80 ml/min per 1.73 m² Ù Delay chemotherapy for 1 week. • 18-23 months <70 ml/min per 1.73 m² • 12-17 months < 65 ml/min per 1.73 m²

Ù Perform estimation of GFR by If no recovery clearance of radioisotope. Ù Use carboplatin 400 mg/m² instead 2 • Age > 2 years, GFR > 60 and < 80 ml/min / m of cisplatin for next course. • Age 18-23 months, GFR > 60 and < 70 ml/min / m 2 Ù Perform estimation of GFR by • Age 12-17 months, GFR > 55 and < 65 ml/min / m 2 clearance of radioisotope before next course . Ù Omit any platinum for next course. • 2 Age > 2 years, GFR < 60 ml/min / m Ù Perform estimation of GFR by • Age 18-23 months, GFR < 60 ml/min / m 2 clearance of radioisotope or urinary • Age 12-17 months, GFR < 55 ml/min / m 2 creatinine clearance before next course.

Page | 99 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

8.3.2 Cisplatin dose modification used in Stratum 3 The normal range for creatinine clearance should be considered to be: • For children over the age of two years GFR should be > 80 ml/min/1.73 m², • For children of 18-23 months of age inclusive GFR should be > 70 ml/min/1.73 m², • For children of 12-17 months of age inclusive GFR should be > 65 ml/min/1.73 m², • For children of 6-11 months of age inclusive GFR should be > 55 ml/min/1.73 m², • For children of 0-5 months of age inclusive GFR should be > 40 ml/min/1.73 m².

The cisplatin dose modification guidelines are the followings:

Serum creatinine > 1.5 x ULN or GFR / Creatinine clearance:

• Age 2 years and over < 80 ml/min per 1.73 m² • 18-23 months <70 ml/min per 1.73 m² Ù Delay chemotherapy for 1 week. • 12-17 months < 65 ml/min per 1.73 m² • 6-11 months < 55 ml/min per 1.73 m² • <6 months < 40 ml/min per 1.73 m²

Ù Perform estimation of GFR by clearance of radioisotope or urinary creatinine clearance. If GFR has not recovered to a point that permits If no recovery the administration of cisplatin then omit cisplatin from that cycle and proceed to the vincristine and carboplatin at the start of the next cycle provided that hematological parameters are satisfactory.

8.4 Ototoxicity of platinum-based chemotherapies

Dose modification of platin (cisplatin or carboplatin) in the event of ototoxicity is based as follows (Note: dose modification is performed based on the highest grade, i.e. the “worst ear”):

Hearing – PTA Dose Modification < 16 dB at 1000 -3000 Hz or None ≤ 40 dB at 4000-8000 Hz 16 -30 dB at 1000 -3000 Hz or Half dose for platin > 40 dB at 4000-8000 Hz > 30 dB at 1000 -3000 Hz and Omit any platin >40 dB at 4000-8000 Hz

Any case of ototoxicity should be discussed with the national committee

Page | 100 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

8.5 Toxicity of methotrexate

Mucositis: In the event of mucositis requiring admission to hospital and intravenous fluids and / or intravenous opiates then the next dose of methotrexate should be reduced by 25%. There should be no subsequent dose escalations. In the event of a further episode of mucositis then the dose of methotrexate should be reduced to 50% with the next cycle of treatment. Delayed excretion of methotrexate: In the event of delayed elimination of methotrexate all possible explanations should be reviewed. These should include effusions or other fluid collections and the medication charts should be reviewed for possible drug interactions. For example, cotrimoxazole should not be given from 7 days prior to methotrexate administration to 7 days following administration of methotrexate because it may increase methotrexate levels by changing protein binding. Any such medications should be omitted in future. If no cause is found then the next dose of methotrexate should be given at full dose, but if there is a second episode of delayed excretion then methotrexate should be omitted from future courses. Nephrotoxicity and delayed excretion of methotrexate: If it has been necessary to use haemofiltration, haemodialysis or glucarpidase, no further administration of high-dose methotrexate is permitted in this protocol.

9 SUPPORTIVE TREATMENTS & CONCOMITANT MEDICATIONS

No other cancer chemotherapy or immunomodulating agents will be allowed during the study. Since all treatment arms contain IMPs that have been used extensively in clinical practice, concomitant medications will be recorded in accordance with regulatory requirements for SAE reporting only. Regarding interactions please refer to the SmPC related to the IMPs used in the country concerned.

Supportive care with antiemetics, antibiotics for suspected fever and neutropenia, and treatment with G-CSF for infection should be given as per local unit policy. Prophylactic treatment with G-CSF is not routinely recommended in strata 1 and 2, but may be used if it is part of local policy. Use of G-CSF is encouraged in stratum 3 to maintain dose intensity if there are delays starting cycles. Hydration fluids and additives may be adjusted according to local practice. The use of dexamethasone as antiemetic therapy should be avoided.

Stratum 2 and 3: Any treatment interfering with MTX elimination should be avoided. The infusion schedule for folinic acid and timings for measuring methotrexate levels are mandatory and must be followed in all cases. In case of extravasation, please follow local guidelines.

Stratum 3: Valproate being an inhibitor of the isoenzymes CYP3A4, CYP2C9 and CYP2C19 and of the cytochrome P450, a special vigilance must be required on the risk of drug interactions.

Page | 101 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

10 ASSESSMENTS

The following assessment will be conducted during the study at different time points as detailed in each study flow chart.

10.1 Medical history, physical and neurological examinations

• Height, weight and surface area, occipitofrontal circumference (at diagnosis and when clinically required) • Assessment of performance status by Lansky (1 year ≤ age < 16 years), or Karnofsky Performance Status (age ≥16 years) • Physical and neurological examination. At staging the neurological assessment should include the Brief Ataxia Rating Scale (BARS) (gait, knee tibia test, finger to nose, dysarthria, oculomotor abnormalities).

10.2 Local and central pathological assessments

Tumour samples preparation and central pathological review At the time of surgery, sufficient fresh material must be addressed to the local pathology department to obtain and store FFPE and frozen tumour samples. • Frozen tumour sample: The on-site pathologist will prepare tissue for usual diagnosis, ensuring at least 1 aliquot is retained for frozen storage. Tissue should be snap-frozen in liquid nitrogen at the earliest opportunity from receipt into the lab. Time from surgery to freezing should be logged and provided to the National coordinator. The frozen tissue is stored at -80°C or in liquid nitrogen, until ready for dispatch to the appropriate centre. If patient consented to the integrated BIOMECA study , frozen material should be sent to the reference BIOMECA laboratory for analyses. • FFPE blocks are prepared as per local centre SOPs. One representative FFPE block used in the diagnostic process must be submitted to the national reference centre. Central pathological review: Tumour tissue delivery to a national referral pathology center is mandatory. According to the national referent pathologist recommendations, either temporary delivery of FFPE block, or at least twenty 5 µm sections on unstained charged slides and at least ten 10 µm curls must be sent. Central review is undertaken according to the World Health Organisation (WHO) Classification 2007. In addition to a more recent revision by SIOP pathologists [21], this will include examination of standard haematoxylin and eosin (H&E) as well as immunohistochemistry to examine other proteins that may be expressed in ependymoma tumours of the posterior fossa and in order to aid and confirm the local centre diagnosis. Pathology specimens will be collected after each surgery.

Lumbar puncture Lumbar puncture is mandatory for all patients. A CSF free of tumour cells is mandatory before randomisation in stratum 1 and stratum 2. It is recommended to perform lumbar puncture 2 weeks after surgery. If a

Page | 102 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

lumbar puncture is performed before 15 days after surgery and is negative for tumour cells, then this will be taken as evidence of non-metastatic disease. If, however, the CSF is positive for tumour cells on lumbar puncture taken before day 15, the lumbar puncture must be repeated at day 15 or later. In case of equivocal CSF cytology, performance of a second lumbar puncture is also recommended. Involvement of CSF pathways by tumour is defined as the unequivocal identification on cytological grounds or with a combination of cytological and immunocytochemical features. According to national regulations a CSF review might be required. Lumbar puncture and metastasis status will be assessed according to the local standard operating procedures of centre. When applicable, CSF should be frozen and stored in local centre ( section 10.10 Integrated biological studies ).

10.3 Biological assessments

10.3.1 Hematology • Full Blood count, Haemoglobin • PT and PTT

10.3.2 Blood chemistry • Serum electrolytes including sodium, potassium, magnesium, phosphate, calcium, • Urea, alkaline phosphatase, alanine transferase or aspartate aminotransferase, bilirubin • Serum creatinine and GFR (calculated creatinine clearance (Ccrea) using Schwartz formulae). Over the first two years of life there is a progressive increase in GFR even when adjusted for surface area. After the second year of life GFR increases in absolute terms, but does not increase when adjusted for surface area. When treating babies with nephrotoxic chemotherapy in the first year of life it is important to adjust the normal range to account for this. Change in GFR over time should be monitored carefully for infants in the first two years of life when treated on Stratum 3. • Serum lactate and ammonium (only for stratum 3)

10.3.3 Urinalysis Urine dipsticks will be used to evaluate the following parameters: • Leukocytes ; nitrite ; protein; blood ; specific gravity; pH

10.4 Radiological assessments

The following MRIs should be obtained preoperatively, postoperatively, upon study requirements and at any time in case of suspected relapse: • Cranial MRI with gadolinium contrast • Spinal MRI with gadolinium contrast (preoperatively, then at relapse only)

Anaesthesia and gadolinium administration should be performed according to the local guidelines. MRIs will be reviewed centrally together with post-operative notes.

Page | 103 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

10.4.1 Cranial MRI The standard imaging plane for the brain should be the axial plane (aligned to the AC-PC axis). Slice thickness should not exceed 4 mm with a slice gap of no more than 0.5mm and must be adapted to each individual location. As the signal of a tumour depends on the field strength of the MRI machine the field strength must not be changed during the study. The identification of a possible very small residual tumour (less than 5 mm in largest diameter) will be included and the exact measurement of such a small structure will necessitate smaller slice thickness than usual. In plane resolution is an essential factor of image quality and therefore a 256 (or 512) matrix for the imaging of the brain and a 512 resolution for spinal canal are necessary. The FOV should be restricted to about 230 mm for the brain and maximum 350 mm for the spinal MRI. The tumour and possible postoperative residue should be measurable in the 3 planes for the calculation of tumour volume (a x b x c/2). 3D-volume calculations may be performed additionally. The a x b x c / 2 volume calculations are the basis for follow-up evaluation and every effort should be made to achieve the greatest achievable accuracy. Sequences for 1-1.5 Tesla MR-machines: • Axial T2 and PD or FLAIR • Coronal FLAIR • Axial T1 -/+ contrast • Coronal and sagittal T1 (-)/+ contrast • Axial DWI with ADC

Optional: 3D gradient echo T1 + contrast and multimodal imaging (e.g. perfusion, MRS, DTI and any other individual local imaging protocols).

For 3 T-MRIs: the T1 imaging usually will be performed as 3D-gradient echo sequence without and with contrast. Additional to the 3D volume sequence T1-SE, gradient echo or T1 inversion recovery images (e.g. in the axial plane) should be acquired.

10.4.2 Spinal MRI Avoid 3T MRI for spinal imaging as the image quality is often inferior to that of 1.5T MR-machines and more unpredictable. The dural sac has to be fully visualized. As only meningeal disease is of interest only sagittal postcontrast T1-weighted sequences are necessary. Slice thickness must not exceed 3 mm. The physiological veins of the cord can be mistaken for nodules of dissemination and therefore axial slices without gaps (SL can be chosen individually) are essential for all suspicious areas. As fat suppression often leads to artefacts and is not necessary for the delineation of meningeal disease it should not be used routinely. T2 TSE sequences or fat suppression techniques are optional.

10.4.3 Early postoperative imaging As non-specific intracranial enhancement is often seen after 3 days following surgery the postoperative MRI must be obtained within this time. If no MRI is performed during surgery, early post-operative MRI must be performed within 72 h, preferentially on day 1 or 2 after surgery. Even within this time false positive nodular enhancement can be seen with haemostatic materials and after electrocoagulation and therefore not only the T1-weighted images before and after contrast need to be evaluated but also the signal intensities on the

Page | 104 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

T2-weighted and FLAIR series. The addition of DWI is valuable for the identification of any post-surgical tissue injury/ischaemia and any associated enhancement. Comparability with the preoperative MRI is essential for the detection of residual tumour. The size of a possible residue has to be measured in the three planes. If the residue is best visible on T2-weighted images a second plane incorporating a T2-weighted sequence must be employed. A residue is considered to be any area of pathological signal and/or enhancement comparable with the appearance of the preoperative tumour. For the evaluation of a residual tumour on imaging the surgical report is often valuable and should be available. It should be noted if spinal MRI is performed postoperatively, non-specific subdural enhancement may be identified on postoperative imaging of the spine and must not be mistaken for meningeal dissemination. In any case of doubt or if intense subdural enhancement is seen the spinal MRI should be repeated within 2 weeks to clarify the situation. Sequences employed should replicate those obtained pre-operatively

10.4.4 Definitions of residual tumour As very subtle residual tumours may not be visible on imaging it is encouraged to compare the results of imaging with the neurosurgical report. A thin line of enhancement can be physiological on early postoperative MRI in the absence of a residual tumour and must not be considered tumour. As patients with residue may be randomized to receive innovative therapy, it is mandatory that imagings are centrally reviewed at a national level or delegates. The residual tumour will be defined as follows (applies only for early postoperative MRI): R0: No residual tumour on postoperative MRI in accordance with the neurosurgical report. R1 : No residual tumour on MRI but description of a small residual tumour by the neurosurgeon or if the neurosurgical result is unknown. R2 : Small residual tumour on MRI with the maximum diameter below 5mm in any direction. R3: Residual tumour that can be measured in 3 planes. R4: Size of the residual tumour not differing from the preoperative status (e.g. after biopsy). RX: If imaging is inadequate or the surgical cavity is very confusing also the term “unclear” should be possible. Every effort should be attempted to clarify the conclusion. Sometimes the presence of blood can be ruled out and distinguished from tumour if the MRI is repeated after some days. Repetition of MRI also may help to distinguish operative changes from residual tumour on T2/FLAIR. After residual tumour assessment by local radiologist, imaging must be centrally reviewed by the national reference radiologist (experienced paediatric neuro radiologist). If the national reference radiologist review differs from the local radiologist evaluation, a web-based or phone teleconference should be organized to expedite resolution of any discrepancies in interpretation of the imaging data. Patients with centrally confirmed R0, R1, R2, may be treated according to standard risk stratum 1. Those with R3 or R4 may be treated with the innovative stratum 2. If despite all efforts, imaging remains RX, patient are not eligible for stratum 1 and 2. If agreed and according to the age, these patients could be recruited either in stratum 3 or in observational study.

Post- operative scans ideally within 24 to 48 hrs of surgery but definitely within 72 hours, imaging should be performed to assess the extent of resection, define the volume of residual disease if present and grade it

Page | 105 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

according to location: Gross total resection (GTR/CR) : surgeon satisfied no solid tumour left (a thin layer of tumour cells on brainstem permissible – if visible only through microscope) – with completely clear scan afterwards with no enhancement. Near total Resection (NTR) : < 5 mm in maximum diameter of residual disease on postoperative scan. Subtotal resection (STR) : > 5 mm of residual disease on post op scan.

10.4.5 Assessment of Response to Treatment Target lesions will be any residual tumour (enhancing or non-enhancing) that has a diameter greater than 5 mm in three dimensions. The tumour and possible postoperative residue should be measurable in the 3 planes for the calculation of tumour volume (a x b x c/2). 3D-volume calculations may be performed additionally. These volume calculations are the basis for follow- up evaluation and every effort should be made to achieve the greatest achievable accuracy. If there is more than one area of residual disease the sum of the products for each lesion will be taken to be the area of residual disease. The tumour response assessment will take into account the radiological assessment (see below), the clinical assessment including the corticosteroids dose: • CCR (Continuous complete response): no evidence of residual or recurrent tumour or meningeal dissemination in a patient previously in complete remission. • CR (complete response): no evidence of residual or recurrent tumour or meningeal dissemination. • PR (partial response): Reduction of tumour volume equal or more than 50% compared to the previous staging MRI. (The development of a meningeal dissemination has to estimated and PR means considerable reduction of meningeal disease) • IMP (improvement or minor response): Reduction of tumour volume between 50% and equal or more than 25% (and minor reduction of a meningeal dissemination) • SD (stable disease): Tumour volume between +25% and -25% compared to the previous staging MRI. (no real change of a meningeal dissemination) • PD (progressive disease): increase of tumour volume of equal or more than 25%, or of appearance of new non target lesion or visual unequivocal increase of non-target lesions or clinical deterioration. An increase in steroid dose alone will not cause a determination of progression in the absence of clinical deterioration or radiological documented lesion growth

Page | 106 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

10.5 Endocrine evaluation

Weight, height and head circumference, Tanner age, early and delayed pubertal onset (breast buds and testes) should be documented at day 0 and at the completion of treatment of each stratum. • TSH, fT4, LH and FSH, oestradiol, testosterone, Insulin-like Growth factor 1 will be measured and analyzed from blood samples collected at day 0 and at the expense of weight gain and/or early puberty • Early Puberty onset (breast buds < 9y, female; 4ml testes <10y male) • Delayed Pubertal onset (>12 y in female, > 13 y in male) • Pubertal Arrest (no pubertal progress, according to Tanner, in one year). • Secondary amenorrhoea of > 3 months or primary amenorrhoea after 13.5 y.

Endocrinological evaluation should be performed according to individual, clinical needs of the patient, at least once per year until age 18. For patients older than 13 years at diagnosis the last evaluation should be performed 5 years post diagnosis. Results of endocrinological assessments need to be documented within the CRF, as these are secondary outcome measures. In case of endocrinological dysfunction, appropriate transition to adult care has to be organized after the surveillance time for the study.

10.6 Audiogram or BAER (Brainstem Auditory Evoked Responses)

Baseline assessment of hearing should be performed at day 0 of each arm containing chemotherapy with analogs of platin. Pure tone, audiometry, air conduction if needed combined with tympanogram, or bone conduction should be performed according to each arm’s specificities described in section 10.12 “Patient follow-up during treatment ”. In case of ototoxicity, please refer to section 8.4 Ototoxicity of platinum-based chemotherapies. For children on stratum 3, audiology should be performed once during each chemotherapy cycle, ideally before cisplatin. As these children are receiving intensive chemotherapy, the timing of audiology cannot be exactly prescribed. The risk of ototoxicity increases with dose so particular vigilance is recommended after cycle 4 of stratum 3. If the child cannot cooperate with pure tone audiometry another age appropriate technique should be used depending on local arrangements. Clinicians may wish to repeat audiometry at more frequent intervals if the child has difficulty in cooperating or concern about hearing is expressed by parents or family.

10.7 Ophthalmology evaluation

Should be performed as required clinically (optional).

Page | 107 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

10.8 Quality of Survival & Quality of life assessment

10.8.1 Quality of Survival: Questionnaires-studies Quality of Survival (QoS) will be indirectly assessed by questionnaires, which are widely used and well accepted by patients and parents. In addition, questionnaires concerning the sociodemographic, educational and vocational status of patients will be added. To facilitate the comparison with the results in further brain tumor studies in Europe (e.g. PNET 5, SIOP CNS GCT II) an identical battery of questionnaires was implemented as listed below. The battery includes questionnaires on health Status, including neuro-cognitive and behavioural outcome as well as on Quality of Life in a closer sense. QoS will be evaluated at baseline and 2 and 5 years post diagnosis as well as at 18 years of age (not necessary in the case of an overlap with the previous named time points). Distribution of questionnaires and measurement at the defined time points will be organized according to national practice.

Health Status, including neuro-cognitive and behavioural outcome: The Health Utilities Index (HUI) [89] is a 15-item questionnaire with 1 additional question that has been shown to be acceptable, reliable and valid in many childhood populations and to be sensitive to clinical problems (excepting behavioral problems) in population of children who have been treated for brain tumors [90]. Please contact Kim Bull - [email protected] or Colin Kennedy - [email protected] for modalities of acquisition.

The Strengths and Difficulties Questionnaire (SDQ) (Goodman, 1999) is a 25-item questionnaire with subscales for hyperactivity, emotional problems, conduct problems, peer relationship and prosocial behavior. Please access the SDQ website ( http://www.sdqinfo.org/ ) for modalities of acquisition of the SDQ.

The Behavior Rating Inventory of Executive Function (BRIEF) parent questionnaire (Gioia 2000) contains 86 items in eight non-overlapping clinical scales and two validity scales. These theoretically and statistically derived scales form two broader Indexes: Behavioral Regulation (three scales) and Meta-cognition (five scales), as well as a Global Executive Composite score. Factor analytic studies and structural equation modeling provide support for the two-factor model of executive functioning as encompassed by the two Indexes. Validity scales measure Negativity and Inconsistency of responses. The BRIEF questionnaire is not provided by the sponsor.

10.8.2 Quality of Life Assessments (in a closer sense) The Pediatric Quality of Life Inventory (PedsQL) [91] and the associated PedsQL Fatigue scale [92] will be used to ask patients under the age of 18 for their subjective rating of physical (eight items), emotional (five items), social (five items) and school (five items) functioning respectively their perception of fatigue. Age- adapted versions for children aged 5-7 years, 8-12 years and 13 to 18 years are available. The PedsQL is a self-administered multidimensional measure of HRQL in healthy children and adolescents and those with acute and chronic health conditions. It consists of 23 items, takes approximately five minutes to complete,

Page | 108 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

and provides information about functioning in four dimensions: physical functioning, emotional functioning, social functioning and school function. PedsQL and the PedsQL fatigue will be provided by the sponsor at national activation in the appropriate language. In place of the PedsQL, adult participants, aged more than 18 years, will complete the EORTC Quality of Life Questionnaire-C30 (EORTC QLQ-C30) [93] Instead of the PedsQL-Fatigue Module, the Multidimensional Fatigue Inventory (MFI) [94] is used to assess the amount of Fatigue in patients aged more than 18 years. Please access the EORTC website ( http://groups.eortc.be/qol/eortc-qlq-c30 ) for modalities of acquisition of the QLQ-C30 questionnaire. MFI questionnaire will be provided by the sponsor at national activation in the appropriate language

Socio-Demographic, Educational and Employment Questions Relevant information on socio-demographic educational and vocational issues should be assessed by the Medical Educational Employment and Social Questionnaire (MEES) [95] widely used and proven in the PNET-protocols. Translations and backward translations in many languages of European countries were provided. The questionnaire needs about 10-15 minutes for completion. The MEES questionnaire is not provided by the sponsor.

Proxy versions and proxy/parent-assessments Proxy versions of all questionnaires are available (except EORTC QLQ C30 and MFI). Parents are invited to complete the proxy versions of all questionnaires for patients younger than 18 years and also the proxy versions of the HUI and the BRIEF for patients older than 18 years as listed in Table 11.

NAME Fill-in by patients Fill-in by parents

HUI ≥ 11 years old patients ≥ 5 years old MEES ≥ 18 years old patients between 5 and 17 years old SDQ 11 - 17 years old patients between 5 and 17 years old EORTC-QLQ-C30 ≥ 18 years old N.A PedsQL 5 - 18 years old patients between 5 and 18 years old PedsQL fatigue 5 - 18 years old patients between 5 and 18 years old MFI ≥ 18 years old N.A BRIEF ≥ 18 years old patients ≥ 5 years old Table 11: Booklets for different age-groups and addresses In participants diagnosed at age 13-18 years, the assessment at age 18 may replace one of the assessments at 2 years or 5 years after diagnosis. For those diagnosed at age 19 or older, this assessment will not be relevant.

Page | 109 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

10.9 Neuropsychological and Neurocognitive assessments

Neurocognitive outcomes will be evaluated by age-appropriate tests. The ‘core’ minimal and US COG compatible assessment battery is presented in Table 12. Standardized cognitive tests will be administered to ensure long-term neurocognitive and learning sequelae are captured. The selected subtests have a long track record in international clinical trials; the battery is time saving and is not expected to cause distress to participants.

Page | 110 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Domain Subtest/s Age range Test Developmental Denver II measure Or Bayley Scales of infant Development III Social-Emotional and Adaptative Behaviour Questionnaire Or Other Developmental Screening test Processing speed Coding 3 -7 years WPPSI -III index Symbol Search + 2 months 6 – 16 years WISC -IV + 11 months 6 – 16 years WISC -III + 11 months 16 -89 years WAIS -IV

Verbal skills Vocabulary 3 -7 years WPPSI -III + 2 months 6 – 16 years WISC -IV + 11 months 6 – 16 years WISC -III + 11 months 16 -89 years WAIS -IV

Fluid intelligence Matrix Reasoning 3 -7 years WPPSI -III/Ravens Raven’s CPM + 2 months 6- 16 years WISC -IV/Ravens + 11 months 6- 16 years WISC -III/Ravens + 11 months 16 -89 years WAIS -IV/Ravens Working CMS Numbers 4-5 years K-ABC/Children’s Memory Scale Memory (forwards & backwards) + 11 months Weschler Digit span 6-16 years WISC -IV (forwards & backwards) + 11 months Digit span 6 – 16 years WISC -III (forwards & backwards) + 11 months 16 -89 years WAIS -IV

Visuo -spatial VMI - Beery -Buktenica / 3- 99 years Beery -Buktenica Developmental Test abilities Drawing - WRAVMA of Visual-Motor Integration/Wide Range Assessment of Visual Motor Abilities (WRAVMA) Reading ability Single Word reading As to national policy/WIAT -II

Motor speed Fine motor Dominant Perdue Pegboard Fine motor Non- WRAVMA dominant Fine motor combined

Table 12: ‘Core’ Neuropsychological Protocol

Page | 111 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

For infants and children aged below the threshold for the locally available Wechsler Preschool and Primary Scale of Intelligence the Bayley III Social-Emotional and Adaptative Behaviour Questionnaire should be utilised. This is suitable for children up to 3 years old. For children 3-6 years please complete ‘core’ tests available on WPPSI III as listed in table 12 where feasible. The ‘core’ neurocognitive assessment battery is the agreed minimum standard required. This was developed to optimise the collection of neurocognitive data from all participating countries. However, participating centers are also encouraged to assess beyond the ‘core’ battery where feasible, using the ‘core plus’ battery (Table 13). This gives opportunity for greater capture, understanding and communication of cognitive and learning effects. Each respective country is encouraged to attach to this protocol, the preferred ‘core plus’ battery assessing the cognitive domains illustrated in table 13, as available and validated in their respective language. Domain Subtest/s Age range Tests Adaptive • Communication 1 day -89 years Vineland II behaviour • Daily Living Skills Vineland II • Socialization • Motor Skills General All core subtests that ensure 2;6 - 7;3 WPPSI -III cognitive index scores are provided. 6- 16 years WISC -III + 11 months 6- 16 years WISC -IV + 11 months 16 - 89 WAIS -IV Memory • Dots 5-16 years Children’s Memory Scale (CMS) • Faces • Stories • Word Pairs • Logical Memory I & II 16+ years Wechsler Memory Scale III (WMS -III) • Faces I & II • Verbal Paired Associates I & II • Family Pictures I & II Executive • Verbal Fluency 8+ years Delis -Kaplan Executive Function Functions • Color-Word System Interference • Trail Making Test <8 years NEPSY Academic • Spelling 5+ years WIAT -II Attainments • Numerical Operations

Attention • Sustained attention Conners’ Continuous Performance Test (CPT) or Test of Everyday Attention for Children (TEA-Ch)* - Score! and Code Transmission

Table 13 : UK ‘Core-Plus’ Neurocognitive Assessments

Page | 112 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Time Points for Assessment : Neurocognitive assessment should take place at the following points where possible: • Diagnosis if possible . It is understood this can be difficult due to tight timescales and other clinical priorities. Clinical judgement should be used to ensure any testing is appropriate and valid. A period of at least 7 days should be allowed after surgery before testing. Testing should take place prior to radiotherapy where possible or within 2 weeks of commencement. • 2 years after diagnosis even if testing at diagnosis was not achieved. • 5 years after diagnosis is also recommended to estimate the long term effects

Administration of neurocognitive assessment and QoS measures detailed in section 10.8 Quality of Survival & Quality of life assessment should be simultaneous. For any issue regarding the neuropsychological and neurocognitive assessments, please contact Dr Sophie THOMAS (Consultant Paediatric Neuropsychologist) – [email protected] .

10.10 Integrated biological study BIOMECA

The SIOP Ependymoma Program II fully supports the identification of informative prognostic biomarkers within the collaborative BIOMECA study. This high priority initiative is an essential element of the overall program to improve future treatment of Ependymoma based on developing a better understanding of the underlying biology. Indeed, biological studies are an important exploratory endpoint of the SIOP Ependymoma II trial Overall we wish to test the hypothesis that key molecular events are predictive of clinical behaviour of ependymoma. • To prospectively evaluate chromosome 1q copy-number status, Tenascin C, H3.3K27me3, RELA-and YAP-fusions as well as and molecular subgroups (by methylation / copy number arrays) as prognostic and predictive biomarkers in ependymoma within clinical trial setting • To identify new biomarkers for clinical and biological behaviour of ependymoma • To compare biology of relapse versus primary disease in patients treated on trial • To prospectively validate the new SIOP ependymoma grading system [21] • To identify and validate circulating markers of ependymoma (CT DNA and micro RNA profiles) based on samples at diagnosis and serially at defined intervals described below

This part of the program will be organised by the European Ependymoma Biology Consortium called Biomarkers of Ependymomas in Children and Adolescents “BIOMECA” throughout a cooperation agreement with the aim to identify informative prognostic biomarkers for the assessment of disease status and predictive response to therapy. The participating site is responsible for ensuring a proper information and consent procedure for patients participating in the biological studies. These consents should include permission to use biological materials for the biological studies purpose.

Page | 113 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

10.10.1 Samples to be collected For all patients participating in the BIOMECA study, a FFPE block must be made available. The investigational site ensures that sufficient material is available for both diagnostic and participation to the BIOMECA study. The FFPE block used to confirm the diagnosis will be transferred by the national referent pathologist to the reference BIOMECA laboratory. We are particularly keen to compare biology of relapse versus primary disease in patients treated on trial and are thus requesting samples from surgery at relapse for biological studies. As surgery is invariably part of the treatment of relapse in ependymoma we highly recommend that samples are made available to BIOMECA so that the underlying biology of relapsed disease can be compared to primary disease and to those tumours that do not relapse when treated according to trial protocol. At the moment and until further notice, additional materials will be asked. These materials are not mandatory for inclusion in the study but highly encouraged: ° Fresh flash frozen remaining tumour tissue by placing tumour in aliquots into cryovials and dropping cryovials (or similar) into liquid nitrogen, ° CSF sample (1-5ml) in sterile universal tubes, spun and then frozen following cytological examination at initial surgery if feasible and at day 15 lumbar puncture (see Appendix 5 for full details), ° Blood sample (5-10ml) in EDTA and gel separator bottles, spun and then frozen for preparation of constitutional DNA(see Appendix 5 for full details), ° Link anonymised copy of local pathology report.

10.10.2 Objectives of the BIOMECA study Biomarkers have been agreed with the BIOMECA consortium (see table 14). In addition, new biomarkers for prognosis and also for specific characteristics of ependymoma with respect to location, metastasis, invasion, recurrence will be screened by using emerging technologies. A series of prospective discovery biomarkers will also be evaluated as part of the BIOMECA program. Mandatory markers include chromosome 1q copy-number status, presence and absence of RELA and YAPfusions, Tenascin-C, H3.3K27me3 and global methylation status. These can all be assessed on FFPE tissue. In this study, in the same way as to what is happening for similar tumours such as medulloblastomas, the best validated biological prognostic criteria, chromosome 1q copy-number status, Tenascin C, Fusion status (RELA, YAP etc) and molecular subgrouping as assessed by methylation profiling will be utilised alongside clinical and histopathological indices, in order to define the low-risk biological profile among ependymomas as well as to identify cases with high-risk biological profiles from the overall cohort. Prospective analysis of the large number of cases will help to: i. Identify clinically relevant disease subgroups, characterize differential genomic signatures; improving our understanding of the molecular basis of ependymoma; ii. Correlate patients response to conventional treatments with specific molecular features in order to identify molecular markers that are valid surrogates of response to treatments; leading to the identification of independent and informative prognostic markers for a more efficient patient stratification in future SIOP clinical trials; iii. Provide experimental evidence to guide the future use of chemotherapeutics as well as novel treatments targeting activated oncogenic pathways; the identification of targets for therapeutic exploitation. iv. To compare biology of relapse versus primary disease in patients treated on trial

Page | 114 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

v. To prospectively validate the new SIOP ependymoma grading system [21] vi. To identify and validate circulating markers of ependymoma (CT DNA and micro RNA profiles) based on samples at diagnosis and serially at defined intervals described below

As a part of the investigative biological studies with newly diagnosed ependymoma patients recruited to SIOP Ependymoma II program, the SIOP Ependymoma Biology Group will be responsible of the validation of already known biomarkers and new biomarkers’ identification. An additional prospective study aim is to define an ependymoma specific signature (CT DNA and micro RNA) that can, in time, determine whether patients have ‘residual molecular disease’ at the end of planned therapy or are in molecular remission. Blood and ideally CSF will be taken at diagnosis, at the end of treatment, and at further specified time points and screened for the presence / absence of an ependymoma specific signature. The patients are followed as normal and the test analysed retrospectively to determine whether the sample might have predicted a relapse or not; a prospective evaluation albeit applied retrospectively. Additional CSF and blood is requested for BIOMECA biological studies when possible for this part of study. CSF (1-5mls) and blood (5mls) sample collection will ideally be taken at the following time points: 1. At diagnosis / surgery (baseline) 2. At the end of treatment Blood samples will then, with patients and parental consent be taken at the following times 3. 1 year after the end of treatment 4. 3 years after the end of treatment 5. or at relapse at any time

An additional blood sample sample 5 years after the end of treatment is also requested Any material surplus to diagnostic needs should in the first instance be made available to the SIOP ependymoma II study in order to provide a sufficient tumour bank to allow identification of key molecular events in ependymoma and to determine prognostic significance of both planned biomarkers and any newly identified markers. The definition of the most relevant ependymoma biomarkers has been made in annual workshops so far taking into account the relevant literature and the progress made by the different research groups participating in the Ependymoma Biology Group. Biomarkers studied and proposal for further testing are following:

Neuropathology The World Health Organization (WHO) 2016 (2) classification defines grades I-III with classic ependymoma (Grade II) and anaplastic ependymoma (Grade III) being more prevalent. A new entity of supratentorial ependymoma has been added defined by the presence of a RELA-fusion gene, comprising 80% of supratentorial ependymomas. Consistent histological grading of ependymoma has proven difficult as a spectrum of pathological features exists and the distinction between Grade II and Grade III is difficult. In turn, the relationship between grade and outcome remains controversial (2,8). An international consensus has proposed new diagnostic features which will be tested in the next SIOP ependymoma trial [21].

Page | 115 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

For further discussion or clarification please contact Professor Richard Grundy Chair BIOMECA consortium

BIOMARKER METHODOLOGY LAB RESPONSIBILITY MANDATORY MARKERS, LEVEL 1 HIGH PRIORITY: 1q GAIN FISH CBTRC MLPA, MIP CBTRC University of Bonn 850k METHYLATION ARRAY CBTRC DKFZ TENASCIN –C (TNC) IHC IGR/St. Anne VIENNA

RELA FUSION RT -PCR University of Bonn IHC Como FISH IGR/St. Anne 850k METHYLATION ARRAY CBTRC DKFZ YAP FUSION RT -PCR University of Bonn IHC Como FISH IGR/St. Anne METHYLATION PATTERN, 850k METHYLATION ARRAY CBTRC COPY NUMBER DKFZ CHANGES H3.3K27me3 IHC CBTRC Bonn/ Vienna MODERATE PRIORITY MARKERS: PI3 KINASE IHC CBTRC IGR ABCB1 & BLBP IHC CBTRC University of Bonn POLYMORPHIC PCR CBTRC VARIATION IN DNA (IGR) REPAIR ENZYME ERCC5 HTERT mRNA RT -PCR Como EZH2 IHC CBTRC IGR CDNK2 MLPA, MIP CBTRC University of Bonn IHC = immunohistochemistry ; FISH = fluorescent in situ hybridization ; RT-PCR = reverse transcription polymerase chain reaction Table 14: Biomarkers studied and proposal for further testing

Page | 116 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

10.11 Valproate PK/PD profile

Pharmacokinetic (PK) and pharmacodynamic (PD) studies will be performed to follow the activity of the valproate in addition to frontline chemotherapy.

• Pharmacokinetic modelling will be carried out using valproate PK parameters in conjunction with patient characteristics and clinical parameters in order to investigate the key factors involved in determining individual valproate drug exposures within the patient population. The PK blood sample analysis will be carried out in the NICR, Newcastle University ( PK sampling is detailed in APPENDIX 6). But, for centres that are not able to support centralised pharmacokinetic studies plasma levels must be determined locally and valproate dosing changed on the basis of local results according to the valproate dosing table below.

• Pharmacodynamics studies will investigate changes in histone H3 and H4 acetylation in fixed circulating leucocytes. Changes between baseline and following achievement of target trough levels of valproate will be correlated with valproate levels and clinical response. The PD blood sample analysis will be carried out in the Northern Institute for Cancer Research (NICR), Newcastle University ( PD sampling is detailed in APPENDIX 6).

10.12 Patient follow-up during treatment

On-treatment assessments will be conducted at different time points as detailed in the study flow charts.

10.13 Patient follow up after treatment (only for interventional strata)

10.13.1 Short and medium term patient follow up after treatment Patients will be followed up to monitor relapse risk till 5 years after completion of treatment. Examinations are listed in table “Follow-up after treatment”.

10.13.2 Long-term patient follow up after treatment Neuroendocrine morbidity, as well as QoS evaluations, will be evaluated until age 18 (or 5 years after diagnosis, if this is after age 18). Endocrine evaluation should be performed according to individual, clinical needs of the patient, at least once per year until age 18. For patients older than 13 years at diagnosis the last evaluation should be performed 5 years post diagnosis. Results of endocrine assessments need to be documented within the CRF, as these are secondary outcome measures. In case of endocrine dysfunction, appropriate transition to adult care has to be organized after the surveillance time for the study.

Other investigations (namely those regarding ototoxicity) will be performed according to local practice but are not required by protocol. If performed and data generated are available, these information could be collected for descriptive analysis.

Page | 117 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Inclusion in Central review Day STAGING staging phase End of Staging Post-operative 15 diagnosis (Within 3 weeks after surgery) procedures confirmation Signature of consent X Medical History (b) and Concomitant Medication X

(h)

Physical Examination and neurological assessment (BARS ) X (a) Weight, Height, occipital circumference X

MRI Brain with gadolinium (To be performed within 72 h, preferentially on day 1 or 2 after surgery if not X(c) performed intraoperatively) MRI Spine with gadolinium X(c) (preferentially preoperatively)

Lumbar puncture (CSF) for local metastasis status evaluation and biological Surgery : Day 0 X (d) (f) study (e) (f)

Blood collection for biological study tissue frozen ideally and FFPE X Central neuropathological review X Central radiological review X Delivery FFPE block to national reference centre for biological study X(g) CSF and blood samples sending to national reference centre for biological study X(f) X Informed consent form for second step of Ependymoma program X

Page | 118 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

STAGING

(a) When surgery is planned, tumour tissue must be sent fresh to the pathology department for preparation for diagnosis and, ideally, 1 aliquot should be temporary retained for frozen storage until consent.

(b) Including personal and familial history of severe epilepsy that should be discarded prior inclusion. Infants or patients ineligible for radiotherapy need to be screened prior to initiation of valproate to exclude the possibility of undiagnosed mitochondrial disorders: persisting elevated blood lactate should be looked prior inclusion into stratum 3 of this program.

(c) Link -anonymized MRI data (imaging and parameters) should be sent for central review in DICOM format on CD or electronically.

(d) Lumbar Puncture should be performed 2 weeks after surgery for cytological examination and biological study (500 µL flash freeze of CSF not needed for cytological diagnosis). If CSF is positive for tumour cells in the lumbar puncture sample taken before day 15, the lumbar puncture must be repeated at day 15 or later. In case of equivocal CSF cytology, performance of a second lumbar puncture is also recommended.

(e) A blood sample should be taken for biological study (5mL, EDTA tube) during a clinically needed blood sampling to avoid extra venipuncture: The blood preparation should be performed ideally within 1 hour after collection.

(f) If available after surgery,biological material should be initially retained at the local treatment centre. Then, samples should periodically be shipped to the national reference centre, which will coordinate the shipment to BIOMECA laboratories.

(g) Tumour tissue delivery to national referral pathology center is mandatory, in the form of at least twenty 4 micron sections on charged slides and at least five 10 µm curls in an eppendorf tube or, preferably, temporary delivery of FFPE tumour tissue block. Whenever possible, collection of frozen tumour tissue, patient’s blood, plasma and cerebrospinal fluid is recommended, upon proper informed consent

(h) BARS parameters: gait, knee tibia test, finger to nose, dysarthria, oculomotor abnormalities

Page | 119 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Stratum 1: Patients with no measurable residual disease and ≥ 12 months of age (Phase III)

Maintenance Completion Day 0 (a) Radiotherapy (b) chemotherapy of

(c) (m) Therapy (d) Inclusion / exclusion criteria review and Signature of Consent for stratum 1 X Concomitant entrytreatments review X Randomisation X Physical Examination and Neurological Assessments X X(e) X(f) X Lansky (1 year ≤ age < 16 years) or KPS (age ≥ 16 years) X X(e) X(f) X Weight X X (f) X QoS/QoL questionnaires and Neuropsychological assessments X MRI Brain with gadolinium X(j) X(k) X Full Blood Count (Diff., Platelets, ANC) + haemoglobin X X(e)

Serum Creatinine or GFR , BUN, Serum Electrolytes (including Ca++, PO4, Mg++) X X(f)

Liver Functions (ALT, AST, PAL, Bilirubin) X X(g) Urinalysis (leukocytes, nitrite, Protein, blood, PH and specific gravity) X Haematology: PT, PTT X Audiogram or BAER (h) X X(h)

Ophthalmology (if clinically required) X Endocrine Evaluation (i) X Blood Pregnancy Test X X(l) CSF (1-5 ml) and blood samples (5 ml) (BIOMECA study) X SAE and AE review X X X X

Page | 120 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Stratum 1: Patients with no measurable residual disease and ≥ 12 months of age (Phase III)

(a) Within 6 weeks after surgery (b) Conformal radiotherapy should take place less than 6 weeks after surgery (c) Maintenance chemotherapy should take place 4 weeks after radiotherapy (d) 3 months after last radiological assessment (e) Weekly (f) Before each perfusion of VEC or CDDP (g) Before each infusion of VEC (h) At D0 and before the third cycle of cisplatin: Pure Tone Audiometry, air conduction if needed combined with tympanogram, or bone conduction should be performed according age of the patient. The clinician may wish to repeat audiometry at more frequent intervals if the patient has difficulty in cooperating or concern about hearing is expressed by him, parents or family (i) TSH, fT4, LH, FSH, Estradiol, testosterone and Insulin-like Growth factor 1 should be measured from blood samples collected at day 0. Also, Tanner age, early or delayed pubertal onset should be documented (j) 6 weeks after the end of conformal radiotherapy if randomized “Radiotherapy + observation” (k) Prior chemotherapy and after cycle 2 and cycle 4 (l) Before the start of chemotherapy (m) For patients under observation , these examinations are facultative and may be performed if necessary according to the opinion of the investigator. In such cases, the data generated from these additional examinations will be collected

Page | 121 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Stratum 2: Patients with inoperable measurable residue and ≥ 12 months of age (Phase II)

Induction Conformal RT Completion Day 0 Maintenance Chemotherapy radiotherapy Boost of (a) Chemotherapy (e) (b) (c) (d) Therapy (f)

Inclusion and exclusion criteria review and Signature of consent for stratum 2 X Concomitant treatments review X Randomisation X Second look surgery discussion X X(p) Physical Examination and neurological assessments X X(i) X(g) X(g) X(j) X Lansky (1 year ≤ age < 16 years) or KPS (age ≥ 16 years) X X(i) X(g) X(g) X(j) X Weight X X(i) X X(j) X QoS/QoL questionnaires and Neuropsychological assessments X MRI Brain with gadolinium X(l) X(n) X(o) X(s) X Full Blood Count (Diff., Platelets, ANC), haemoglobin X X(g) X(h) X(h) X(g) Serum Creatinine or GFR , BUN, Serum Electrolytes (including Ca++, PO4, Mg++) X X(i) X(j) Liver Functions (ALT, AST, PAL, Bilirubin) X X(i) X(q) Methotrexate Dosage X(m) Urinalysis (leukocytes, Nitrite, Protein, blood, specific gravity, PH) X Haematology: PT, PTT X Audiogram or BAER X X(k) Ophthalmology (if clinically required) X Endocrine Evaluation (t) X Blood Pregnancy Test X X(r) CSF (1-5 ml) and blood samples (5 ml) (BIOMECA study) X SAE and AE review X X X X X X

Page | 122 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Stratum 2: Patients with inoperable measurable residue and ≥ 12 months of age (Phase II)

(a) Within 3 weeks after surgery. (b) Induction chemotherapy (VEC +/- HD-MTX) should take place ideally 3 weeks after surgery. (c) Conformal radiotherapy should take place as soon as possible (before 6 weeks after the last infusion of induction chemotherapy or 6 weeks after second look surgery). (d) Radiation boost should take place immediately after conformal radiotherapy if confirmation that radiation boost is needed by MRI analysis (performed at the end of conformal radiotherapy). (e) Except in patients with progression of disease during frontline chemotherapy with VEC +/- HD-MTX. This maintenance chemotherapy should be started 4 weeks after the completion of radiotherapy. (f) 3 months after last radiological assessment (g) Weekly (h) If required. (i) Before each perfusion of VEC or MTX (D1, D15, D22, D36, D43 et D57 of induction chemotherapy with HD-MTX or D1,D22 and D43 of induction chemotherapy without HD- MTX) (j) Before each perfusion of VEC or CDDP (D1, D21, D36, D57, D71, D85 and D106 of maintenance chemotherapy). (k) Before the third cycle of cisplatin (Day 57 of maintenance chemotherapy). (l) 3 weeks after the last infusion of induction chemotherapy: Imaging and parameters should be sent as soon as available to National central review committee to assess second look surgery possibility. This exam is necessary for the primary endpoint evaluation. (m) After each MTX course : MTX level should be measured at 48h, 72h, 96h after the last start of the infusion and a 24h intervals thereafter until level is < 0.1µmol/ L (n) During the last week of conformal radiotherapy to the tumour bed, a new diagnostic MRI and a planned CT must be obtained for all patients. Imaging and parameters should be sent as soon as available to the National central review committee to evaluate need for cRT boost . (o) Four weeks after the second radiation fraction of the conformal radiotherapy boost and just before maintenance chemotherapy. This exam is necessary for the evaluation of boost toxicity. (p) Discussion on second look surgery should take place at the end of the induction chemotherapy. (q) Before each infusion of VEC. (r) Before the start of maintenance chemotherapy (s) After cycle 2 and cycle 4 (t) TSH, fT4, LH, FSH, Estradiol, testosterone and Insulin-like Growth factor 1 should be measured from blood samples collected at day 0. Also, Tanner age, early or delayed pubertal onset should be documented

Page | 123 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Stratum 3: Randomized phase II chemotherapy study in children < 12 months or those not eligible to receive radiotherapy (phase II) Chemotherapy Completion of Day 0 (a) +/- valproate (b) chemotherapy (c) Inclusion and exclusion criteria review and Signature of consent for stratum 3 X Ammoniemia and lactic acid blood level X Randomisation X Physical examination and neurological assessments X X (d) X Lansky (1 year ≤ age < 16 years) or KPS (age ≥ 16 years) X X (d) X Denver, QoS/QoL and Neuropsychological assessments according to age X Weight and Height X X(d) X MRI Brain with gadolinium X(e) X MRI Spine with gadolinium X (f) X Full Blood Count (Diff., Platelets, ANC), haemoglobin X X(d) Serum Creatinine , BUN, Serum Electrolytes (including Ca++, PO4, Mg++) X X(d) Glomerular filtration X X(g) Liver Functions (ALT, AST, PAL, Bilirubin) X X(d) Urinalysis (leukocytes, Nitrite, Protein, blood, specific gravity, PH) X Haematology: PT, PTT X Audiogram or BAER X X(h) Ophthalmology Evaluation (if clinically required) X Endocrine Evaluation (m) X Methotrexate dosage X(i) Pharmacodynamics (Optional): HAC3 & HAC4 in PBMC (ELISA) (j) X X Pharmacokinetics: PK sampling (k) Or local assessment for trough valproate measurements (l) X X CSF (1-5 ml) and blood samples (5 ml) (BIOMECA study) X SAE and AE review X X X

Page | 124 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Stratum 3: Randomized phase II chemotherapy study in children < 12 months or those not eligible to receive radiotherapy

(a) Within 3 weeks after surgery (b) Ideally within 3 weeks after surgery

(c) 3 months after last radiological assessment.

(d) Before first infusion of each chemotherapy cycle.

(e) Brain MRI with gadolinium should be performed at the end of cycle 2, 4 and 6 of the chemotherapy.

(f) If initially positive

(g) Before first infusion of Cycle 2, 5, 6 and 7 of chemotherapy.

(h) Once during each chemotherapy cycle, ideally before cisplatin.

(i) After each MTX course : Methotrexate level should be measured at 48h, 72h, 96h after the last start of the infusion and a 24h intervals thereafter until level is < 0.1µmol/ L

(j) Optional: Blood samples (2ml) will be obtained pre-treatment and at 4h post dose administration. Additional samples to be taken following achievement of steady state trough valproate levels and at 6 monthly intervals thereafter. Tubes will be stored at -20°C once mixed with fixative (as detailed in lab manual) prior to transport to the Northern Institute for Cancer Research (NICR), Newcastle University.

(k) Optional: Blood samples (2ml) will be obtained pre-treatment and at 1, 2, 4, 6 and 12h after the initial valproate dose and at the same time points following valproate dose escalation where carried out. An additional trough sample will be taken at 12h following valproate administration once the targeted valproate concentration has been maintained for a period of two weeks (steady state). All samples will be taken from a central venous line. Blood samples will be immediately transferred to heparinized tubes, centrifuged for 5 min at 2,000 rpm and 4°C and the plasma transferred to a clean labeled tube and stored at -20°C prior to transport to the Northern Institute for Cancer Research (NICR), Newcastle University.

(l) Mandatory: for centers that are not able to support pharmacodynamic study we suggest that plasma levels will be assessed locally and valproate dosing changed on the basis of local results according to the valproate dosing table presented in this protocol (see section Valproate ).

(m) TSH, fT4, LH, FSH, Estradiol, testosterone and Insulin-like Growth factor 1 should be measured from blood samples collected at day 0. Also, Tanner age, early or delayed pubertal onset should be documented

Page | 125 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

FOLLOW UP AFTER 2nd year after From 3 rd to 5 th year First year after From 5 years after completion of completion of after completion of TREATMENT completion treatment treatment to 18 years of age (f) (only interventional strata) treatment treatment

Physical Examination Full neurological examination Every 3 months Every 4 months Every 6 months local practice (g) Lansky or Karnofsky Tanner age, pubertal onset and occipital circumference At 1 year At 2 years Once per year Once per year (h)(i) Endocrine evaluation: TSH, LH, FSH, fT4, GH (a)

Hearing Function (b) At 6 months and at 1 year none at 5 years local practice (g)

MRI Brain with gadolinium (c) Every 3 months Every 4 months Every 6 months (c) local practice (g) MRI Spine with gadolinium (d)

QoL & QoS assessments none At 2 years (l) At 5 years (l) At 18 years old (j) (k) Neuropsychological assessments At least one per Renal function (e) At least once At least one per year local practice (g) year

Blood Chemistry : Serum urea, electrolytes, creatinine and Once at 1 year none Once at 5 years local practice (g) magnesium

Optional blood sample for BIOMECA At 1 year none At 3 years none study (m)

Page | 126 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

FOLLOW UP AFTER TREATMENT

(a) Should be documented by a pediatric endocrinologist or pediatrician with endocrine expertise (b) Should be assessed by pure tone audiometry (c) Children enrolled in the stratum 3 should have local MRI scans at 3 monthly intervals until 2 years after the end of treatment and then 4 monthly to 5 years (d) If initially positive for infants and patients not eligible to radiotherapy (patients in stratum 3). (e) If chemotherapy was performed. (f) Updating OS and PFS data may be requested beyond the 5th year of follow up for patients. (g) Should be performed according to local practice but are not required by protocol. If performed and data generated are available, these information could be collected for descriptive analysis. (h) For patients older than 13 years at diagnosis the last evaluation should be performed 5 years post diagnosis. (i) In case of endocrine dysfunction, appropriate transition to adult care has to be organized after the surveillance time for the study. (j) Neuropsychological morbidity will be evaluated only until 5 years after diagnosis. (k) QoS/QoL will be evaluated until age 18 (or 5 years after diagnosis, if this is after age 18) (l) QoS/QoL and neuropsychology will be evaluated at 2 years and 5 years after diagnosis (m) Blood samples (5 ml) should be collected at 1 year and 3 years after the end of treatment, or at relapse at any time.

Page | 127 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

11 PATIENT WITHDRAWAL

11.1 Withdrawal from trial treatment The reason for withdrawal from protocol treatment should be recorded in the patient’s medical notes and on the case report form (CRF) where it is due to either the patient’s or clinician’s decision. Reason for withdrawal may include, but are not limited to: ° The patient withdraws consent to further trial treatment, ° The patient was subsequently found not be eligible, ° Unacceptable toxicity, ° Disease progression whilst on therapy.

The strata will be analysed on an “intention-to-treat” basis and any patients withdrawn from trial treatment will remain within the trial unless the patient explicitly withdraws consent for data collection.

11.2 Withdrawal of consent to data collection

If a patient or parents / guardians explicitly states that they do not wish for any further data to be collected and/or existing data to be analysed this must be recorded on the relevant CRF. Details should also be recorded in the patient’s hospital records and no further CRFs must be completed.

12 CLINICAL SAFETY DATA MANAGEMENT

12.1 Generalities The Investigator should assess the seriousness and causality of all AEs experienced by the patient (this should be documented in the source data) with reference to the relevant Summary of Product Characteristics.

12.1.1 Definitions (Directive 2001/20/EC of the European Parliament) Adverse Event (AE): An adverse event is any untoward medical occurrence in a patient administered a medicinal product or medical therapy, which does not necessarily have a causal relationship with this treatment. An AE can therefore be any unfavourable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of a medical therapy, whether or not related to the medical therapy. In order to monitor the safety of all trial participants throughout their trial participation this definition has been extended to include any treatment provided in the trial (e.g. radiotherapy). The legal definition does not need extension to untoward medical occurrences before consent signature, because any procedures at that time are standard procedures for these patients and do not subject them to any trial specific additional risk.

Page | 128 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Adverse Reaction (AR): All untoward and unintended responses to an investigational product to any dose administered. The definition covers also medication errors and uses outside what is foreseen in the protocol. The definition implies a reasonable possibility of a causal relationship between the event and the investigational medicinal product. This means that there are facts (evidence) or arguments to suggest a causal relationship. ‘No reasonable possibility’ means that firstly the time relationship to drug administration is improbable (with the knowledge at the time), and/or another explanation is more likely.

Unexpected Adverse Reaction : An unexpected adverse reaction is an AR the nature or severity of which is not consistent with the applicable drug information. Reports which add significant information on the specificity, increase of occurrence, or severity of a known, already documented serious adverse reaction constitute unexpected reactions. Examples of UARs: • Unexpected outcome (e.g. fatal) of an expected AR (=SUSAR) • Increase in the rate of occurrence of an expected AR, which is judged to be clinically important, is considered as unexpected • New report of more specific disease (e.g. interstitial nephritis) instead of a labeled, more general

Serious Adverse Event (SAE) or Serious Adverse Reaction (SAR): Any untoward medicinal occurrence or effect that at any dose • results in death • is life-threatening • requires hospitalization or prolongation (of at least 1 day) of existing hospitalization • results in persistent or significant disability or incapacity (1) • pregnancy • is a congenital anomaly, foetal malformation or abortion • or is medically significant (2)

(1) The terms disability and incapacity refer to any clinically significant physical or mental handicap, whether temporary or permanent, that affects the patient’s physical activity and/or quality of life. (2) A medically significant event is any clinical event or laboratory test result considered by the investigator to be serious and that does not correspond to the seriousness criteria defined above but may jeopardize the patient or require medical intervention to prevent one of the serious outcomes mentioned in the definition above (for example, intensive treatment in an emergency room or at home for allergic bronchospasm, blood dyscrasias or convulsions that do not result in hospitalization, overdose, second cancer, pregnancy and new events may be considered medically significant).

Suspected Unexpected Serious Adverse Reaction (SUSAR): Any serious event that is not mentioned or differing in terms of nature, intensity, evolution (clinical course) or frequency from that listed in the Summary Product Characteristics (SPCs).

Page | 129 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

12.1.2 Intensity criteria The intensity criteria must not be confused with the seriousness criterion asked when declaring a SAE. The Cancer Therapy Evaluation Program (CTEP) Version 4.03 of the NCI Common Terminology Criteria for Adverse Events (CTCAE) will be used for intensity grading of reported AEs (Appendix 3 – TOXICITY Criteria of Adverse Events v4.0 3 (CTCAE ). The intensity of adverse events not listed in this classification will be assessed using the following descriptors: • Mild (grade 1): does not affect the patient’s usual daily activities, • Moderate (grade 2): disturbs the patient’s usual daily activities, • Severe (grade 3): prevents the patient’s usual daily activities, • Life-Threatening (grade 4): requires critical care / is life-threatening, • Death (grade 5).

12.1.3 New safety issues Any new data that could lead to revaluate the ratio between the benefits and risks of the research, or that could be sufficiently important to consider modifications of the research documents, the research management and the drug utilization or to suspend, to interrupt or to modify the protocol of the research or of the similar researches.

12.2 Reporting Requirements

12.2.1 Events that require immediate reporting on a Serious Adverse Event Form All AEs that meet the definition of an SAE (see section 12.1.1 ) should be reported immediately by investigator through a SAE Form (except those described in section 12.2.2)

12.2.2 Events that do not require an immediate reporting on a SAE form Except for valproate, the use of all investigational drugs is current standard of care for paediatric patients with ependymoma and the design of strata are experimental only regarding to the treatment schedule dosing and not in respect of the use of the substances. The safety profile of these drugs is well established. Exceptions to immediate SAE reporting obligations for certain expected or anticipated events therefore do not compromise patients' safety. All SAE which do not require immediate reporting will be documented on the CRF forms. This applies from first day of trial treatment until end of the follow-up period defined by the protocol. The investigators’ SAE reporting obligations to the sponsor are thereby fulfilled. All SAE, including those not requiring immediate reporting, are regularly seen by the Data Monitoring and Safety Committee. ° Expected hospitalisation for procedures such as blood transfusion for haematological toxicity CTCAE grade 1-4 without complications, antibiotic treatment of neutropenia fever or CTCAE grade 1-3 infections, nutritional support for weight loss CTCAE grade 1-3 or other expected toxicity CTCAE grade 1-3 (e.g. bleeding or haematuria due to thrombocytopenia) is not to be immediately reported on an SAE form. Moreover, it will be documented on the respective therapy / toxicity CRF. ° Hospitalisation due to signs and symptoms of disease progression as long as the outcome does not lead to death during protocol treatment and for 30 days after last protocol therapy is not to be immediately reported on an SAE form. It will be documented on the respective CRF for documentation of relapse / progression.

Page | 130 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

At regular intervals, the sponsor may ask to report all these events (e.g. at the time of preparing interim safety reports for the iDSMC).

12.2.3 Events that do not require SAE report ° Any hospitalisation for the treatment of the patient’s cancer or its complications,Any hospitalisation for the present research ° Any hospitalisation planned prior to study entry, ° Any hospitalisation for pre-existing conditions that have not been exacerbated by trial treatment, ° Any hospitalisation to simplify treatment procedures ° Any hospitalisation of less than 24h (emergency) or when no complication is observed, ° Cancer-related death, occurring after the notification period (30 days after withdrawal). The non-serious adverse events which would be related to these conditions will be reported in the appropriate sections of the patient’s CRF.

12.2.4 Pregnancies A female participant must be instructed to stop taking the study drug and immediately inform the investigator if she becomes pregnant during the study. A positive urine test must be confirmed by a serum pregnancy test. If it is confirmed that the patient is not pregnant, the patient must resume dosing with study drug, providing they agree to use adequate contraception for the remainder of the trial. The investigator should report all pregnancies within 24 hours to the sponsor, using the “SAE Report Form”. The investigator should counsel the subject; discuss the risks of continuing with the pregnancy and the possible effects on the foetus. Monitoring of the patient should continue until conclusion of the pregnancy. Pregnancies occurring up to 7 months after the completion of the study medication must also be reported to the investigator. Pregnancy occurring in the partner of a male participant in the study should be reported to the investigator and the sponsor. The partner should be counselled, the risks of continuing the pregnancy discussed, as well as the possible effects on the foetus. Monitoring of the patient should continue until conclusion of the pregnancy. The investigator must notify the Sponsor of the outcome of the pregnancy as a follow-up to the initial SAE report. It is important to monitor the outcome of pregnancies of partner in order to provide SAE data on congenital anomalies or birth defects.

In the event that a patient or his partner becomes pregnant during the SAE reporting period please complete a Serious Adverse Event report form (providing the patient’s details) and return to the Trials Office as soon as possible. If it is the patient who is pregnant, provide outcome data on a follow-up Serious Adverse Event report form. Where the patient’s partner is pregnant, consent must first be obtained and the patient should be given a pregnancy release of information to give to his partner. If the partner is happy to provide information on the outcome of her pregnancy this information should be indicated in the patient’s medical file. Once consent has been obtained and documented, provide details of the outcome of the Serious Adverse Event report form.

Page | 131 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

12.2.5 Reporting periods During staging phase: SAE will be documented and reported from the date of Staging ICF signature and up to 30 days after lumbar puncture. During interventional strata: SAE will be documented and reported from the date of ICF signature and up to 30 days after the last study drug administration or later if investigator states that there is a reasonable causal relationship with study drugs. During observational study: No tolerance data will be reported.

12.3 Reporting Procedure This section refers to Adverse Events occurring in patients from one of the interventional strata. Adverse Events occurring during the observational study may not be reported into the eCRF .

12.3.1 Adverse Events AEs should be reported on the AE Form (and where applicable on an SAE Form). AEs will be reviewed at each visit using the Common Terminology Criteria for Adverse Events (CTCAE), version 4.03.

Any AEs experienced by the patient but not included in the CTCAE should be graded by an Investigator and recorded on the AE Form using a scale of (1) mild, (2) moderate or (3) severe. For each sign/symptom, the highest grade observed since the last visit should be recorded.

For each stratum, all adverse events grade ≥3 regardless of relationship to the treatment are to be recorded on the corresponding pages included in the case report form (CRF). For each AE occurred, only the higher grade achieved between two visits and its duration (days) will be recorded in the eCRF. Whenever possible, a diagnosis, rather than a symptom should be provided. Laboratory values, as well as vital signs and ECG abnormalities are to be recorded as Adverse Events only if they are considered medically relevant by the investigator: symptomatic, requiring corrective treatment, leading to discontinuation/dose modification (reduction and/or delay), and/or fulfilling a seriousness criterion. Signs and symptoms that are present prior to the first study drug administration are to be recorded on the medical history pages included in the CRF. They are to be recorded as adverse events as soon as one of the following changes occurs: increased intensity, relationship, action taken regarding study drug. AE related to HD-MTX (stratum 2) and valproate (stratum 3) will be followed-up until complete resolution or return to baseline level.

12.3.2 Serious Adverse Events All Serious Adverse Events (SAE) or new safety issue, expected or unexpected, whether related or not to the study treatments or procedures, occurring during the study from the date of signed informed consent till 30 days after last study drug intake, must be reported by the investigator to the pharmacovigilance without delay (or immediately) of being aware of the event. Pregnancy must be reported as soon as it is confirmed.

Page | 132 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

The investigator will have to complete the “SAE Report Form” (available in the Investigator Master File) immediately or no later than 24 hours after the investigator becomes aware of the event. The form must be dated, signed by the investigator and sent to the sponsor: Either by fax: Fax: +33 (0)9 81 40 42 80 (pharmacovigilance) and Fax: +33 (0)4 78 78 27 15 (European coordinating center)

Or by e-mail: E-mail: [email protected]

For each event, the investigator will report: ° The patient’s identification number (in the study), ° A reporter name and signature, ° A diagnosis and the medical description of the SAE or any symptoms (as accurate as possible and using medical terminology) and any other information which could provide important information on the event, ° The date of onset and recovery for the event, ° The intensity of event (grading according to NCI CTC AE v4.03), ° The name of the investigational products administered, ° The seriousness criterion, ° The investigator statement of causality (related/unrelated) of the event with the study drug, ° The action taken (no action taken; study drug dosage adjusted/temporarily interrupted; study drug permanently discontinued due to this AE), ° Concomitant medications / therapies used, ° The clinical outcome of event. If the event was not fatal, it should be monitored until recovery, until the patient has returned to his/her previous condition, or until any sequelae have stabilized, ° The patient’s relevant medical history.

Whenever possible, the investigator should also provide any additional document necessary for SAE assessment, such as: ° A copy of the hospitalization or prolonged hospitalization report, ° A copy of the autopsy report, if applicable, ° A copy of all the results of any additional tests performed, including relevant negative results, along with the normal laboratory values, ° Any other document that is relevant in the investigator’s opinion. All these documents must be anonymised. The investigator shall supply the sponsor with any additional requested information. More information might be requested by the pharmacovigilance or the study CRA through a SAE Query Form. In case of incomplete information, the investigator will have to provide follow-up information as soon as possible, again using the “SAE Report Form” (within 3 days). The investigator is responsible for providing appropriate medical follow-up for patients until resolution or stabilization of the AE or until the patient’s death. If required, follow-up might be extended beyond patient’s withdrawal or end of study.

Page | 133 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

The original SAE Report Form will be archived by the investigator in the appropriate Investigator Master File section. A copy of CIOMS is stored by the sponsor in the trial master file.

12.3.3 Reporting to the Competent Authority and main Research Ethics Committee Sponsor duties The sponsor (Centre Léon Bérard) is undertaking the duties of trial for the reporting of safety events during the whole study. For all SAEs, the sponsor will evaluate the causal relationship to the study drug, or to the study protocol and the expectedness of all SAEs based on knowledge of the adverse reaction on the reference document: the SmPCs. The causality assessment given by the local Investigator cannot be overruled, and in case of disagreement, both opinions will be provided in any subsequent reports.

The sponsor will have to report any suspected unexpected serious adverse reactions (SUSAR) to the EMA (Eudravigilance database), competent authorities and Ethics Committees in the concerned Member States if required by local laws, and Independent Data Safety and Monitoring Committee (iDSMC). SUSAR must be reported: • without delay after first knowledge for fatal or life-threatening SUSARs, • as soon as possible but no later than 15 calendar days after first knowledge for all other SUSARs. In each case, relevant follow-up information should be sought and a report completed as soon as possible and communicated to EMA, the Competent Authorities and Ethics Committees in the concerned Member States, if required by local laws within an additional 8 calendar days for all SUSARs. The sponsor keeps detailed records of all adverse events. The sponsor provides to Competent Authorities of the Member States in whose territory the clinical trial is being conducted, the Ethic Committees and Independent Data Safety and Monitoring Committee a Development Safety Update Report (DSUR), listing all suspected serious adverse reactions (expected and unexpected) and all serious adverse events which have occurred over this period. The sponsor alerts national study coordinators in case of new safety issue identification and proposes urgent safety measures to be applied. The sponsor provides without delay to Competent Authorities of the Member States in whose territory the clinical trial is being conducted and to the Ethic Committees with any declaration of new safety issue, as well as possible measures to be proposed discussed with the coordinator. Additional relevant information concerning the New Safety Issues must be reported within 8 days.

The Sponsor shall notify the Principal Investigators of any information which might affect the patient’s safety.

According to the agreement between the sponsor and the participating member states, the clinical safety data management may be delegated to an identified local structure as in charge of the PV in the country concerned. All safety data will be then centralized by the pharmacovigilance of the sponsor.

Page | 134 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

13 DATA HANDLING AND RECORD KEEPING

13.1 Database and Data Collection

There will be a single international study database held by the biostatistics Unit of the Sponsor (DRCI of Centre Leon Bérard). All the data concerning the patients will be recorded in the eCRF throughout the study. SAE reporting will be entirely paper-based. Data entry will be performed online by investigators and authorised staff. A print out of the electronic sheet must be available at any stage of the study. A final copy of the eCRF of each patient will be signed by the Investigator and filed at the Centre. All the data will be computerised to detect missing, out of range and inconsistent data. Self-evident corrections will be performed by the data manager and documented according to a validated data management plan. Queries will be made for wrong and missing data. Data Clarification Forms (DCFs) will be generated in the eCRF; investigators or its authorized staff will have to solve the query in compliance with source data. A patient eCRF will be validated after no more inconsistency is detected by the program. All database modifications will be recorded in the audit trail. The database will be submitted to a quality control. The database will be locked after all queries are solved, and after data review and final validation. The sponsor is the owner of all collected data and reports.

13.2 Archiving

Personal patient data will be kept confidential. Each investigator will keep in his file a screening log and a patient identification list (including complete name and date of birth of each patient). The Patient/Parent Information Sheet and the informed consent form have to explain that the study data will be kept in an electronic form, confidentially, in accordance with the French law N°78-17. Study documents and records will be kept for at least 15 years after the end of the study by the investigator: this includes codes allowing the identification of patients and all the source data, the eCRF print out, the queries, the patient's informed consent, the copies of competent authorities’ approval and any other document pertinent to the study. All other documentation relative to the trial (protocol, investigators’ folders, etc…) as well as all other study specific documents (laboratory results, radiographic pictures, reports of physician-patient consultations and physical examinations, etc…) are confidential material and must be kept in a secured location.

Page | 135 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

14 QUALITY MANAGEMENT

14.1 Site Set-up and Initiation All sites will be required to sign appropriate contracts with their National Coordinating Centre prior to participation. In addition all participating Investigators will be asked to sign the necessary agreements and supply a current CV with evidence of recent GCP training to the National Coordinating Centre. All members of the site research team will also be required to sign a Site Signature and Delegation Log which lists the range of duties that have been delegated to them for the trial. This should be counter-signed by the Principal Investigator and a copy should be returned to the National Coordinating Centre. Prior to commencing recruitment all sites will undergo a process of initiation. Key members of the site research team will be required to attend either a meeting or a teleconference covering aspects of the trial design, protocol procedures, Adverse Event reporting, collection and reporting of data, and record keeping. Sites will be provided with an Investigator Site File and a Pharmacy File containing essential documentation, instructions, and other documentation required for the conduct of the trial. The National Coordinating Centre must be informed immediately of any change in the site research team.

14.2 Monitoring

This trial will be monitored in accordance with the ICH Note for Guidance on Good Clinical Practice (ICH Topic E6, 1996). Monitoring visits to the trial sites will be organised by the National Coordinator and conducted on behalf of the Sponsor. Monitoring visits will allow analysing the study progress, to check the existence of signed informed consent, the compliance of patients with inclusion and non-inclusion criteria and serious adverse events source documents and all data related to endpoints assessment. The clinical research assistant will have to check the compliance of eCRF data with source documents, as specified in the GCP guidelines. They will ask the investigator to modify any wrong or forgotten data. All modifications will be explained (if necessary). If data are modified by another person than the investigator, the authorisation of this person will be documented on the Site Signature and Delegation Log. If needed, the sponsor shall implement the centralized review of certain protocol criteria. Any major problems identified during monitoring may be reported to the Trial Management Group, Trial Steering Committee and the relevant regulatory bodies.

Onsite Monitoring Monitoring will be carried out according to the Trial Monitoring Plan. It is the responsibility of participating National Coordinating Centres to ensure that the level and process of monitoring described in the Monitoring Plan is in accordance with their national regulations and to inform the Sponsor of any issues.

Additional on-site monitoring visits may be triggered for example by poor CRF completion rates, poor data quality, low or high SAE reporting rates, excessive number of patient withdrawals or deviations. If a monitoring visit is required the National Coordinating Centre will contact the site to arrange a date for the proposed visit and will provide the site with written confirmation. Investigators will allow the SIOP Ependymoma II Program staff access to source documents as requested.

Page | 136 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

The Patient/Parent Information Sheet and the informed consent form will specify that the Sponsor representative and the Competent Authorities will have direct access upon request to the source documents related to the study, including medical antecedents, to check the accuracy of the data.

Central Monitoring Trial staff from the Sponsor will be in regular contact with the National Coordinating Centre/site research team to check on progress and address any queries that they may have. Trial staff will check incoming CRFs for compliance with the protocol, data consistency, missing data and timing.

National Coordinating Centres/Sites may be suspended from further recruitment in the event of serious and persistent non-compliance with the protocol and/or GCP. Any major problems, including serious breaches of GCP and/or the trial protocol identified during monitoring may be reported to the Sponsor and to the national coordinating center that is in charge of reporting to relevant regulatory bodies.

14.3 Audit and Inspection

The Investigator will permit trial-related monitoring, audits, ethical review, and regulatory inspection(s) at their site, providing direct access to source data/documents. Health Authorities must be permitted to inspect all trial-related documents and other materials at the site, including the Investigator Site File, the completed CRFs, the IMP(s), and the subjects’ original medical records/files. Audits may be conducted at any time during or after the trial to ensure the validity and integrity of the trial data. Sites are also requested to notify the Sponsor and the National Coordinating Centre of any CA inspections.

14.4 Notification of Serious Breaches

This trial is sponsored by Centre Leon Bérard, in France. In accordance with Regulation 29A of the Medicines for Human Use (Clinical Trials) Regulations 2004 and its amendments the Sponsor of the trial is responsible for notifying the licensing authority in writing of any serious breach of: • The conditions and principles of GCP in connection with that trial or; • The protocol relating to that trial, within 7 days of becoming aware of that breach For the purposes of this regulation, a “serious breach” is a breach which is likely to effect to a significant degree: • The safety or physical or mental integrity of the subjects of the trial; or • The scientific value of the trial National Coordinating Centres are therefore requested to notify the Sponsor of a suspected trial-related serious breach of GCP and/or the trial protocol. Where the Sponsor is investigating whether or not a serious breach has occurred sites are also requested to cooperate in providing sufficient information to report the breach to the CA where required and in undertaking any corrective and/or preventive action.

Page | 137 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

15 END OF TRIAL DEFINITION For regulatory purposes, the end of each stratum will take place after the last active patient has completed follow up. This will allow sufficient time for the completion of protocol procedures, data collection and data input. The Sponsor will notify the Competent Authorities(s) and Ethic Committee(s) that the trial has ended at the appropriate time and will provide them with a summary of the clinical trial report within 12 months of the end of trial.

16 STATISTICAL CONSIDERATIONS

16.1 Overall program Descriptive statistics will be produced for GTR rate and second look surgery rate. Based on the results observed, an updating of OS and PFS data may be requested by the steering committee beyond study duration.

16.2 Treatment Stratum 1: Patients with no measurable residual disease and ≥ 12 months of age (Phase III)

16.2.1 Sample size calculation The primary endpoint of this study is progression free survival (PFS). The study is designed to detect an improvement of the 5-year PFS rate of 15% with the chemotherapy arm as compared to the observation arm (75% vs 60%). Assuming 10 years for recruitment, 13 years for total study duration (i.e 3 years for the last included patient), and two interim analyses for futility, 160 patients per arm (109 events in total) are required with a power of 85% and a bilateral alpha risk of 5 %.

16.2.2 Disease Stratification Randomisation will be stratified on: ° Site (supra vs infratentorial) ° Grading (Classic vs anaplastic) ° Age (More or less than 3 year old at diagnosis)

16.2.3 Analysis populations Efficacy analyses will be performed on the Intent-to-treat population (ITT) including all randomised patients. Safety analysis will be performed on the safety population including all patients having received at least one dose of study treatment.

16.2.4 Statistical methods Primary endpoint analysis Progression-free survival (PFS) will be measured from the date of randomisation to the date of event, defined as progression or death due to any cause. Patients with no event as the time of the analysis will be censored at their last adequate tumour assessment. PFS will be estimated by the Kaplan-Meier method. The

Page | 138 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

5-year PFS rate will be described per treatment arm with its 95% confidence interval. PFS distribution will be compared between the 2 arms using the Log-Rank test.

Secondary endpoints analysis Overall survival (OS) will be measured from the date of randomisation to the date of death due to any cause and will be censored at the date of last follow-up for patients alive at their last follow-up. OS will be estimated using the Kaplan-Meier method and will be compared between the 2 arms using the Log-Rank test. Descriptive statistics will be used to evaluate neurocognitive, neuroendocrine late effects and QoS in each treatment arm. The assessment of safety will be based mainly on the frequency of adverse events based on the common toxicity criteria (CTC-AE-V4.03) grade. In each study arm, descriptive statistics will be provided for characterizing and assessing patient tolerance to treatment. Adverse events will be coded according to the MedDRA®.

16.2.5 Planned interim analysis and stopping guidelines Two interim analyses are planned during the study, after observation of 37 events for the first analysis and 74 events for the second one. At each interim analysis, efficacy and safety data will be examined by an independent Data Safety Monitoring Committee (iDSMC). From a safety point of view there are no pre-defined rules but according to the benefit-risk ratio observed a recommendation to continue or not the study with or without amendment will be proposed to the steering committee. To help in the decision, the iDSMC will have the possibility to refer to pre-defined futility stopping rules. Using a beta spending function approach (Lan and DeMets, 1983) with O’Brien-Fleming boundaries, these indicative rules will recommend stopping the trial earlier for futility if : • At the first interim analysis, the p-value comparing the 5-year PFS rate between the 2 treatment arms is ≥ 0.98 • At the second interim analysis (in case of no early stopping before), the p-value comparing the 5- year PFS rate between the 2 treatment arms is ≥ 0.38

16.2.6 Planned final analysis The final analysis of trial data will be performed after 3 years follow-up of the last included patient.

16.3 Treatment stratum 2: Patients with inoperable measurable residue and ≥ 12 months of age (Phase II)

16.3.1 Sample size considerations for the randomised phase II treatment stratum This trial has been designed as a single stage randomised phase II Jung design. Due to the rarity of the condition, the sample size is limited by availability of patient numbers. It is anticipated that in the ten years of study recruitment, it will be possible to randomise at least 60 patients - 30 patients into each arm. With 30 patients per arm randomised, under the assumption that the true response rates are 30% in the control arm and 45% in the methotrexate arm then the suggested analysis rule below will give 79% power to successfully carry methotrexate forward for further evaluation, but with a 34% risk that it will be carried forward when in fact it is ineffective.

Page | 139 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

16.3.2 Statistical methods for the randomised phase II treatment stratum Due to the small number of patients in the study there will be no formal hypothesis testing. Point estimates of treatment effects together with 95% confidence intervals will be produced.

Primary outcome measure Primary analyses will be by the ITT principle. The number of treatment responses on each arm will be counted and the difference between them will be calculated. Assuming 30 patients are recruited per arm, and then the suggestion will be to take methotrexate forward for further evaluation if this difference is ≥2 in favour of methotrexate. This gives the power and false positive rates above. This rule is not intended to provide a prescriptive trial result. Instead, the decision to take methotrexate forward will be based on a more pragmatic assessment including assessment of toxicity. The difference in number of treatment responses will be interpreted as per the table below:

Number of patients per arm Analysis rule 30 35 40 Take forward if 2 79% / 34% 83% / 35% 85% / 36% difference is ≥ __ 3 71% / 24% 75% / 26% 79% / 27% in favour of Methotrexate 4 61% / 16% 67% / 18% 72% / 20%

The cell values give the power / false positive rate An estimate of the treatment effect will be calculated as an odds ratio and 95% confidence interval. However this is expected to be wide due to the small number of patients in the trial.

Secondary outcome measures Survival data will be analysed by the log rank test. Mean hazard ratios and 95% confidence intervals will be given. Dichotomous data will be analysed as per the primary outcome.

16.3.3 Planned subgroup analysis No subgroup analyses are planned, due to the small number of patients in the trial.

16.3.4 Planned interim analysis and stopping guidelines An interim analysis of safety and activity will be carried out annually and the results will be presented to an independent Data Safety and Monitoring Committee (iDSMC).

Methotrexate High Dose: There will be no formal stopping rules applied, but it is anticipated that a difference of at least 3 standard errors in favour of methotrexate in the primary outcome would need to be observed to justify early termination of the study for efficacy.

Radiotherapy boost Monitoring of key safety endpoint, i.e. toxicity, will be conducted continuously. The table reported below provides decision rules as to the continuation or not of accrual to the trial; it will serve as a trigger for

Page | 140 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

consultation with the iDSMC for additional review, and does not represent formal “stopping rules” that would mandate automatic closure of study enrolment. The table was generated according to the methodology of continuous monitoring for toxicity using Pocock-type boundary)[96] with the following assumptions: toxicity rate of 7%, 5% probability of early stopping, sample size N of 60. The trial is stopped if the number of toxic events is equal or larger than the boundary for a given number of patients evaluable for toxicity. For instance, based on the figures shown in the table the trial would be stopped if, over the first 20 evaluable patients, toxicity is observed in 5 or more.

N.of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 patients Boundary - 2 3 3 3 3 3 4 4 4 4 4 4 4 5 5 5 5 5 5

N.of 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 patients Boundary 5 6 6 6 6 6 6 6 6 7 7 7 7 7 7 7 7 8 8 8

N.of 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 patients

Boundary 8 8 8 8 8 8 9 9 9 9 9 9 9 9 9 10 10 10 10 10

Boundary is equivalent testing the null hypothesis, after each patient, that the event rate is equal to 7%, using a one-sided level 0.013899 test.

Radionecrosis from grade 2 MRI documented, with neurologic impairment, requiring steroids and not improving after 6 months of treatment is considered a toxicity case and stopping criterion.

If radionecrosis occurred, close and iterative contacts between local principal investigator and the chief investigator responsible of the Boost Study (Italian Team) must be implemented in the best delay.

16.3.5 Planned main analysis The first main analysis of trial data will take place 6 months after the final patient has been entered into the study, when response data from all patients will be available. A further analysis will be carried out when more follow up data is available for the survival outcomes.

Page | 141 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

16.4 Treatment stratum 3: Randomised phase II chemotherapy study in children < 12 months of age or those not eligible to receive radiotherapy (Phase II)

16.4.1 Sample size considerations The study aims to recruit a minimum of 100 patients over 10 years, with 2.5 years follow up at the end of the study. Median follow up will therefore be 5 years. Assuming that the true 5 year PFS rates are 45% and 60% in the control and valproate arms respectively then the study will have 80% power to detect the difference between the arms at a 1-tailed alpha level of 0.25. If recruitment is less than target, and only 82 patients are recruited, then the study retains 75% power to detect this difference. Conversely, if the study manages to recruit 130 patients then it will have 80% power to detect the difference at a 1 tailed alpha level of 0.2 .

16.4.2 Statistical methods Primary outcome measure Primary analyses will be by the ITT principle. The primary outcome measure will be analysed by means of the log rank test. As this is a phase II study assessing the merit of taking valproate forward to phase III, the study is not powered to demonstrate a clear treatment effect. The alpha level of the test will therefore be one sided and relaxed, so that addition of valproate will be considered worthy of further investigation if it proves statistically significantly better than standard care at a one-tailed p value of 0.25 or less. Due to the lack of power present in this study, likelihood Bayesian methods will be used to construct and interpret a posterior distribution of the range of possible treatment effects (given as hazard ratios), assuming a non-informative prior.

Secondary outcome measures Survival outcomes will be analysed as per the primary outcome. For toxicity, proportions of patients experiencing grade 3/4 toxicity will be compared between arms.

16.4.3 Planned subgroup analyses There are no planned subgroup analyses.

16.4.4 Planned interim analysis and stopping guidelines An interim analysis of safety and efficacy will be carried out annually and the results will be presented to the iDSMC. There will be no formal stopping rules applied, but it is anticipated that a difference of at least 3 standard errors in favour of valproate in the primary outcome would need to be observed to justify early termination of the study for activity.

16.4.5 Planned main analysis The first main analysis of trial data for primary outcome will take place 2.5 years after the final patient has been entered into the study. This does not preclude the possibility of further future analyses of long term follow up data.

Page | 142 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

17 TRIAL ORGANISATION AND STRUCTURE

17.1 Sponsor

The Centre Leon Bérard is the Sponsor for this study.

17.2 Coordinating Centres

According to GCP guidelines, each cooperating National Coordinating Centre is required to determine a National Coordinating Investigator, who is a corresponding member of the international SIOP Ependymoma II study committee. The National Coordinating Investigator must be qualified for this position as required by the national laws and regulations of the respective country.

This includes the certification of all cooperating hospitals and may include patient insurance.

While patients are under therapy, frequent contacts (phone or e-mail) will be held between the participating institutions, the National Coordinators and the Sponsor to discuss the clinical status of the trial.

17.3 Relationship of the Trial Committees

Sponsor

Progress Reports Questions & Feedback

iDSMC: iDSMC TSC feedback & TSC: Trial independent feedback to respond to iDSMC TSC via TMG Steering Data Safety and TMG comments Reports sent to Committee Monitoring Reports plus iDSMC iDSMC feedback Committee

17.3.1 Trial Management Group The Trial Management group (TMG) is nation specific and is composed of the Chief Investigator, Co- investigators and the trial team of the relevant National Co-ordinating Centre. The TMG is responsible for the day-to-day running and management of the trial and will meet by teleconference or in person as required.

Page | 143 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

17.3.2 Trial Steering Committee A steering committee will be composed of Sponsor representatives (project manager and statistician manager), the principal investigator, and representatives of main investigators involved in the study. Meetings via teleconference will be scheduled as deemed appropriate from the beginning of the trial. Additional meetings may be called at any time upon request of any members. On an ongoing basis, the Steering Committee will review all Grade ≥ 3 adverse events and Critical Events. Upon request, it will evaluate: • Potential amendment • Statistical analysis plan • Results of interim safety analyses • Study continuation and possibility to recruit additional patients (increase of the required sample size) in case a lot of patients will be unevaluable before the primary endpoint evaluation.

The steering committee is also empowered to propose the inclusion (or non-inclusion) of any participating center. It will be regularly informed of the accrual rate of inclusion and of any emergent problems and will review the activity and safety data at the end of the study.

Based on the results observed, a retrospective updating of OS and PFS data may be requested by the steering committee of the beyond study duration.

17.3.3 Independent Data Safety and Monitoring Committee (iDSMC) An independent Data Safety and Monitoring Committee (iDSMC) will be set-up prior to the start of the trial. The iDSMC will be constituted by the Sponsor and will be composed of three international experts: methodologists, statisticians and medical experts. Each of these members must be unconnected with the trial and cannot, therefore, be one of the trial investigators. The role of the iDSMC will be to monitor the accrual rate, to examine interim analyses and to monitor toxicities. In case of unexpected safety concerns, the iDSMC will liaise with the Steering committee of the program for assessment of benefit and risk of the trial. The iDSMC will operate in accordance with appropriate regulatory guidelines. Data analyses will be supplied in confidence the iDSMC, which will be asked to give advice on whether the accumulated data from the trial, together with the results from other relevant research, justifies the continuing recruitment of further patients. The iDSMC will also be informed on a real time of all severe adverse event or critical event, and of any emergent problems by the Sponsor. This committee will meet to review each event that could modify the benefit risk ratio of the study (i.e. scientific, safety, ethical events). Additional meetings may be called at any time if an event occurs or on request by one or more members. The iDSMC may consider recommending the discontinuation of the trial if the recruitment rate or data quality are unacceptable or if any issues are identified which may compromise patient safety. The iDSMC plays an advisory role for the Sponsor, who has the final decision regarding the implementation of recommendations put forward by the iDSMC. The Sponsor is responsible for sending iDSMC recommendations to the regulatory authorities. (For stopping rules please refer to statistical considerations section ) If appropriate, iDSMC reports and study progress reports will be prepared every 6 months for the first 2 years of the study, and later upon iDSMC approval will be generated once every year if the study goes smoothly during the first 2 years.

Page | 144 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

18 FINANCIAL CONSIDERATIONS

This is an investigator-initiated and investigator-led trial. It is the responsibility of the National Coordinating Centres to obtain the necessary funding. The Sponsor will also seek of European funding to cover administrative costs of the program.

19 ETHICAL CONSIDERATIONS

The accepted basis for the conduct of clinical trials in humans is founded on the protection of human rights and the dignity of human beings with regard to the application of biology and medicine and requires the compliance with the principles of Good Clinical Practice (GCP) and detailed guidelines in line with those principles (Directive 2001/20/EC (2) and Directive 2005/28/EC (1)). GCP (Good Clinical Practice) is a set of internationally recognised ethical and scientific quality requirements which must be observed for designing, conducting, recording and reporting clinical trials that involve the participation of human subjects. Compliance with good clinical practice provides assurance that the rights, safety and well-being of trial patients are protected, and that the results of the clinical trials are credible (Article 1 (2) of Directive 2001/20/EC).

The Sponsor and Investigators shall consider all relevant guidance with respect to commencing and conducting a clinical trial (Article 4 of the Directive 2005/28/EC) and take into account the consensus on harmonisation for GCP by the International Conference on Harmonisation – ICH (Directive 2005/28/EC (8)).

The conduct of this trial shall be based on the following international ethical and statutory source: the 18 th World Medical Association Declaration of Helsinki , Finland, June 1964, amended at the 48th World Medical Association General Assembly, Somerset West, Republic of South Africa, October 1996 (website: http://www.wma.net/en/30publications/10policies/b3/index.html ).

If the region has adopted the Convention for the protection of Human Rights and dignity of the human being with regard to the application of biology and medicine: Convention on Human Rights and Biomedicine (CETS No.: 164)

(Council of Europe – Ratification signed in the following countries: Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Georgia, Greece, Hungary, Iceland, Italy, Latvia, Lithuania, Luxembourg, Moldova, Montenegro, Netherlands, Norway, Portugal, Romania, San Marino, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, The Former Yugoslav Republic of Macedonia, Turkey, Ukraine).

Directive 2001/20/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 4 April 2001 on the approximation of the laws, regulations and administrative provisions of the Member States relating to the implementation of good clinical practice in the conduct of clinical trials on medicinal products for human use (Official Journal L21, 01/05/2001 P. 0034 – 0044) and detailed guidances.

Page | 145 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Directive 2005/28/EC of 8 April 2005 laying down principles and detailed guidelines for good clinical practice as regards investigational medicinal products for human use, as well as the requirements for authorisation of the manufacturing or importation of such products (Official Journal L 91, 09/04/2005 P. 0013 – 0019).

Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data (Official Journal L 281 , 23/11/1995 P. 0031 – 0050).

Moreover, in each participating member state, the trial will be conducted in accordance with the national clinical trial regulations and laws in force including medicine for human use, data protection, Human tissue collection and transportation.

Scientific guidelines relating to the quality, safety and efficacy of medicinal products for human use, as agreed upon by the CHMP and published by the Agency, as well as the other pharmaceutical Community guidelines published by the Commission in the different volumes of the rules governing medicinal products in the European Community (as stated in the Directive 2005/28/EC (9)).

The protocol will be submitted to and approved by the National Research Ethics Committee (REC)s and Competent Authorities before the study can open in each respective country.

Before patients can be registered on to the study, each site must obtain all necessary regulatory approvals in accordance with their national laws. It is the Principal Investigators responsibility to ensure any subsequent amendments gain the necessary approval. This does not affect the clinicians’ responsibility to take immediate action if thought necessary to protect the health and interest of individual patients.

20 CONFIDENTIALITY AND DATA PROTECTION

Personal data recorded on all documents will be regarded as strictly confidential and will be handled and stored in accordance with the appropriate laws in the participating national state i.e. National Data Protection Laws. With the patient’s consent, their partial date of birth, hospital identifying number, general/medical practitioner details will be collected at trial entry to assist with long-term follow-up via other health care professionals. Patients will be identified using only their unique registration number, initials, hospital number and partial date of birth on the Case Report Form and correspondence between the Trials Office and the participating site. Where appropriate, patients are asked to give permission for the Trials Office to be sent a copy of their signed Informed Consent Form which will not be anonymised. This will be used to perform in-house monitoring of the consent process. The Investigator must maintain documents for submission to the National Co-ordinating Centre or Sponsor (e.g. Patient Identification Logs) in strict confidence. In the case of specific issues and/or queries from the regulatory authorities, it will be necessary to allow access to the complete trial records, provided that patient confidentiality is protected.

Page | 146 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

The National coordinating centre and Sponsor will maintain the confidentiality of all patients’ data and will not disclose information by which patients may be identified to any third party other than those directly involved in the treatment of the patient and organisations for which the patient has given explicit consent for data transfer. Representatives of the SIOP Ependymoma Program trial team may be required to have access to patient’s notes for quality assurance purposes but patients should be reassured that their confidentiality will be respected at all times.

21 INSURANCE

This is an investigator-initiated and investigator-led trial. Therefore, the Sponsor will delegate to National Coordinating Centres the insurance subscription for SIOP Ependymoma II Program in their respective countries and for ensuring that sites in their country are adequately covered.

22 PUBLICATION POLICY

Intellectual property rights and publication policy will be addressed in the corresponding contracts between Sponsor and National Coordinating Centres/sites.

Page | 147 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

APPENDIX 1 – INTERNATIONAL CONTACT DETAILS

WRITING COMMITTEE

Dr Martin ENGLISH Dr Didier FRAPPAZ Dr Lorenza GANDOLA Hanane GHEIT Pr Richard GRUNDY Dr Maura MASSIMINO Dr Piergorgio MODENA Dr Geraldina POGGI Dr Katja VON HOFF Pr Keith WHEATLEY

Page | 148 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

APPENDIX 2 – WMA DECLARATION OF HELSINKI

Ethical Principles for Medical Research Involving Human Subjects:

http://www.wma.net/en/30publications/10policies/b3/17c.pdf

Page | 149 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

APPENDIX 3 – TOXICITY CRITERIA OF ADVERSE EVENTS V 4.03 (CTCAE)

Toxicities will be recorded according to the Common Terminology Criteria for Adverse Events (CTCAE V4.03). The full CTCAE document is available on the National Cancer Institute (NCI) website, the following address was correct when this version of the protocol was approved: http://ctep.cancer.gov/protocolDevelopment/electronic_applications/ctc.htm

Page | 150 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

APPENDIX 4 – SUMMARY OF PRODUCT CHARACTERISTICS

The current version is available on the MHRA drug database website: http://www.mhra.gov.uk/spc-pil/?subsName=

Page | 151 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

APPENDIX 5 – BIOMARKERS OF EPENDYMOMA IN CHILDHOOD AND ADOLESCENCE (BIOMECA)

The aims of the collaborative BIOMECA consortium are: • To identify new biomarkers for Ependymoma clinical and biological behaviour (location, recurrence, chemo resistance, invasion, metastasis). • To validate known biomarkers in the context of international prospective study for newly diagnosed Ependymoma • To validate and compare the techniques to assess biomarkers for further use in the stratification of patient • To select the best biomarker(s) and establish a prognostic signature for ependymoma • Discovery and exploration of novel biomarkers and therapeutic targets • To compare biology of relapse versus primary disease in patients treated on trial • To prospectively validate the new SIOP ependymoma grading system [21] • To identify and validate circulating markers of ependymoma (CT DNA and micro RNA profiles) Several candidate biomarkers have been proposed. However markers showing reproducible results in prospective and sizeable groups of young ependymoma patients treated in a homogenous manner within the context of a clinical trial are still required. Indeed, several purported biologic prognostic markers in ependymoma have been shown to lose this capacity when assessed across clinical trial cohorts, highlighting the importance of analysing any prospective marker in standardised therapeutic groups. To date four biomarkers have been evaluated in retrospective cohorts of Childhood ependymoma in training and test sets and are sufficiently robust to include in this study as mandatory markers; chromosome 1q gain, Tenascin –C. Fusion transcripts (RELA & YAP) and copy number change alterations. A number of prospective discovery biomarkers need to be evaluated as part of the BIOMECA program (see section BIOMECA study ). Tumour samples guidelines The primary purpose of the surgical operation is to debulk the tumour aiming for complete resection where possible. The neurosurgeon should aim to ensure that maximal amount of tissue is sent to the pathology department of the investigating centre concerned. Priority is to ensure that sufficient material is available for the local neuro pathologist to make a primary diagnosis. Any material surplus to diagnostic needs should in the first instance be made available to the SIOP ependymoma II study in order to provide a sufficient tumour bank to allow: o Central pathological review performed by the national reference pathologist o The Ependymoma grading according the SIOP grading scheme by the national reference pathologists to allow the harmonization of the ependymoma grading which has been developed by SIOP reference Neuro pathologists [21] o The BIOMECA Study ( by BIOMECA reference laboratory: Identification of key molecular events in ependymoma and to determine prognostic significance of both planned biomarkers, any newly identified markers) and for future biological research.

Page | 152 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Sample Requirements for SIOP Ependymoma II:

1. Formalin Fix sufficient sample for diagnostic and research purposes (MANDATORY for inclusion in the study) 2. Copy of local treatment centre pathology report 3. Flash-freeze remaining tumour tissue by placing tumour in aliquots into cryovials and dropping cryovial (or similar) into liquid nitrogen. PLEASE NOTE FROZEN TISSUE IS HIGHLY RECOMMENDED and will be audited. 4. Blood sample 5-10mls in EDTA and gel separator bottles – Spun and frozen for preparation of constitutional DNA (see BLOOD AND CSF SAMPLES FOR ANALYSIS OF CIRCUALTING TUMOUR NUCLEIC ACIDS) 5. Cerebrospinal fluid – Spun and frozen following cytological examination (see BLOOD AND CSF SAMPLES FOR ANALYSIS OF CIRCUALTING TUMOUR NUCLEIC ACIDS) 6. As surgery is invariably part of the treatment of relapse in ependymoma we highly recommend that samples are made available to BIOMECA so that the underlying biology of relapsed disease can be compared to primary disease and to those tumours that do not relapse when treated according to trial protocol

Please note sufficient tissue will always be left in any FFPE blocks for further diagnostic use .

Material surplus to biological studies requirements will be returned to local neuropathology laboratory according to best practice in that participating country.

Page | 153 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

The Local Centre

CONSENT Once the pathologist has confirmed the tumour is ependymoma, it is the local centre's responsibility to undertake consent procedures as per their national SOPs to allow participation in the clinical trials, and to permit investigative biological studies to be undertaken on the material on a link-anonymised basis. These consents should include permission to use other materials (e.g. blood, cerebrospinal fluid samples; see separate blood and CSF collection below ) taken in investigative biological studies. Consent must be obtained early by the clinical team to allow rapid trial inclusion and submission of tissue to BIOMECA reference laboratory. Please notice that: • Formalin Fixed Tumour Tissue for central neuropathological review is a Mandatory part of the protocol • Fresh tumoral tissue, blood and CSF collection are not mandatory for inclusion but are highly encouraged.

TUMOUR SAMPLES When surgery is planned for a paediatric Central Nervous system tumour, tissue must be sent fresh to the local pathology department to confirming the ependymoma diagnosis and allow optimal tissue banking to support inbuilt biological studies. If possible (not mandatory but highly encouraged), please freeze (at -80ºC or below) as much tumour tissue as possible (frozen tissue will be used by BIOMECA in order to extract high quality DNA and RNA for high resolution genomic analyses to test the hypothesis that key molecular events are predictive of clinical behaviour of ependymoma) as well as providing a FFPE block or sections on charged slides in order to validate prospectively the new SIOP histological grading scheme within BIOMECA reference laboratory. Where possible additional FFPE blocks or sections on charged slides could be provided for BIOMECA studies

BLOOD AND CSF SAMPLES FOR ANALYSIS OF CIRCUALTING TUMOUR NUCLEIC ACIDS

Blood acquisition and local preparation Blood should be acquired for: - Extract constitutional DNA as control material for genomic analyses performed on the frozen tumour tissue - Harvest the plasma AND serum for research into markers of minimal residual disease / circulating tumour nucleic acids.

Blood samples of 5-10ml should be taken at day 0 and at future time intervals as described. These should be linked to blood sampling needed for clinical purposes to avoid extra venepunctures:

1) BLOOD COLLECTION FOR PLASMA SAMPLES a) Collect 5-10ml blood into an EDTA blood tube labelled with the patient’s trial number and the date taken. b) Gently invert tube 8-10 times and stand blood tube in an upright position at room temperature. c) Within 1 hour after collection, centrifuge samples at 1600g for 10mins at room temperature.

Page | 154 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

d) Transfer 1ml aliquots of the plasma (top layer) from the blood tubes into labelled 1.5ml sterile Eppendorf tubes (supplied by CBTRC on request), being very careful to avoid contamination of the buffy coat (middle layer) or red blood cells (bottom layer). e) Transfer 0.5ml aliquots of the buffy coat (middle layer) from the blood tubes into labelled 1.5ml sterile Eppendorf tubes and store at -80 oC. Discard remaining sample left inside the blood tube according to local rules. f) Store all aliquots at -80°C at the local centre until transfer of all samples is arranged to the UK BIOMECA laboratory.

2) BLOOD COLLECTION FOR SERUM SAMPLES a) Collect 5-10ml blood into a yellow topped Vacutainer® gel separator blood tube (or if unavailable, use a red topped Vacutainer) labelled with the patient’s trial number and the date the taken. b) Stand the blood tube in an upright position for 30 minutes at room temperature. c) Centrifuge at 2000g for 10mins at room temperature. d) Transfer 0.5ml aliquots of the serum (top layer) from the blood tubes into labelled 1.5ml sterile Eppendorf tubes (supplied by CBTRC on request) and place all aliquots into a -80°C freezer. Discard remaining sample left inside the blood tube according to local rules. e) Store aliquots at -80°C at the local centre until transfer is arranged to the paired BIOMECA laboratory.

CSF acquisition and local preparation CSF should be obtained at surgery (day 0) as a baseline sample if possible in addition to CSF at mandatory day 15 lumbar puncture. We recommend a minimum volume of 1 ml but ideally 5 to 10 ml to be processed and sent to the paired BIOMECA laboratories (see details below).

1) CSF SAMPLE PROCESSING a) Collect approximately 5-10ml CSF where possible into a sterile container labelled with the patient’s trial number and the date the CSF was taken (1ml minimum). b) Centrifuge CSF at 180g for 10 minutes. c) Remove the supernatant to a freshly labelled tube leaving the “invisible” cell pellet in the original container. d) Place tubes (CSF supernatant and cell pellet) into a -80°C freezer. e) Store tubes at -80°C at the local centre until transfer is arranged to the paired BIOMECA laboratory (for onwards transfer of all samples to UK BIOMECA centre).

It is allowed (and recommended for keeping time lines for stratification and to start radiotherapy, if the clinical situation allows lumbar puncture) to perform lumbar puncture at day 0 if possible.

A lumbar puncture will be performed 2 weeks after surgery. If this is negative for tumour cells, then this will be taken as evidence of non-metastatic disease. If, however, the CSF is positive for tumour cells on lumbar puncture taken before day 15, the lumbar puncture must be repeated at day 15 or later. In case of equivocal CSF cytology, performance of a second lumbar puncture is also recommended.

Page | 155 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

In addition to the CSF material routinely collected for local cytological assessment, additional CSF should be harvested when possible.

Shipment to BIOMECA reference laboratory After preparation, frozen Tumour, FFPE research blocks, Blood and CSF samples will be retained at the investigating centre concerned and stored as indicated. Samples should periodically be shipped to the BIOMECA reference laboratory. The local centre will liaise to arrange submission to the BIOMECA reference laboratory of blood, CSF samples (after local clinical cytology assessment), as well as any relevant documentation for transport according the national operating arrangements. Receipt of material will be confirmed by the BIOMECA reference laboratory

The local pathologist

The local pathologist will prepare tissue for usual diagnosis, ensuring at least 1 aliquot is retained for frozen storage. Priority is to ensure that sufficient material is available for the local pathologist to make a primary diagnosis.

Tissue should be snap-frozen in liquid nitrogen at the earliest opportunity from receipt into the lab. The frozen tissue is stored at -80°C or in liquid nitrogen, until ready for despatch to the national reference laboratory. The approximate time between surgical resection and tissue freezing should be recorded.

The Local centre will contact the national coordinating centre and the BIOMECA centre that has been agreed by National group immediately once consent and ependymoma diagnosis is locally confirmed. The national coordinating centre will liaise to arrange rapid submission of diagnostic material and the BIOMECA centre will advise on transfer of frozen tissue, FFPE, Blood and CSF (in some countries the National Reference Pathology and BIOMECA are co-located) .

Receipt of material will be confirmed by the national reference and BIOMECA laboratories.

A Link-anonymised copy of local pathology report as well as any Link-anonymised relevant documentation for transport according national operating arrangements. Receipt of material will be confirmed by the national reference pathologist and by BIOMECA centre co-ordinator.

Local pathologist will be kindly asked to:

Ù send to national reference pathologist: one FFPE tumour tissue block and H&E’s from all FFPE blocks or Twenty 5µm FFPE sections on charged slides with sufficient interpretable material. Where possible any local immunohistochemistry should also be submitted to the national reference pathologist for pathology review to avoid duplication of staining. The immunohistochemical slides will be returned to the local pathologist as soon as the review is complete. Please, Note that this blocks should be the same block that used for local diagnosis and will always returned to local centre with residual tissue.

Page | 156 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Ù Prepare for BIOMECA laboratory: At least one formalin fixed paraffin embedded research block. The number of blocks to be sent varies according tumours material available ( please refer to the tumour sample management flow chart) – FFPE blocks should be sent with diagnostic FFPE block to national reference pathologist for preparation before sending to BIOMECA reference laboratory.

Fresh frozen tumour is not mandatory for inclusion but highly encouraged (Flash freeze remaining tissue by placing in aliquots into cryovials and dropping cryovials (or similar) into liquid nitrogen).

The national reference pathologist

From the 20 x 5µm clinical diagnostic FFPE block sections or blocks and slides sent by the local pathologist:

• To confirm local diagnosis: the national reference pathologist will perform a central pathological review according to the latest revision of the World Health Organisation (WHO) classification for tumours of the CNS, currently the version from 2016.

To validate prospectively the new histological grading scheme from SIOP: a harmonized approach on grading will be used developed by SIOP reference Neuropathologists [21]. This new histological grading scheme has been proposed and needs to be validated prospectively. It takes into account the cell density, the mitotic activity and the microvascular proliferation. Necrosis and atypia are not considered anymore for grading to prospectively assess histopathological features of the samples and explore the relevance of this new ependymoma grading scheme. This will include examination of standard haematoxylin and eosin (H&E) as well as immunohistochemistry to look at markers such as expression of neuroepithelial proteins including, GFAP, and epithelial membrane antigen (EMA), as well as proteins that may be expressed in other tumours of the posterior fossa, e.g. synaptophysin, neurofilament protein, cytokeratins, Olig-2, LIN28, INI1 to aid in confirming the local centre diagnosis. Ki-67 immunohistochemistry will be included as a marker of cellular turnover. The aim of this study is the prospective application of the algorithm and further optimization for the histopathological diagnostics of paediatric ependymoma.

The following will be assessed in each case by a panel of central review neuropathologists blinded to outcome. ° Mitotic count (per 10 HPF) ° Hypercellularity (% or area) ° Vascular proliferation ° Necrosis ° WHO grade ° Ependymal rosettes / channels ° Ki67 / MIB1 (% positive cells)

Individual features as well as the deduced new SIOP classification will then be tested against outcome parameters

Page | 157 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

If Research FFPE sections or block(s) are available: • to prepare material necessary to BIOMECA reference laboratory in order to build a tissue micro- array: A 5 µm section will be will be cut from this FFPE block and stained with H&E. This will then be marked up by the national reference pathologist, to indicate regions of exclusive tumour cell content which are representative of the entire sample this will allow tissue micro-array construction by BIOMECA reference laboratory with which your participating country has paired,

• Using the marked-up H&E for reference, up to 5 x 1.0 mm tissue cores will be taken from the donor FFPE tissue blocks submitted and transferred to a recipient composite block. Tissue core positions will be robustly transcribed. The number of cores taken will depend upon the surface area and depth of the donor FFPE block. PLEASE NOTE: sufficient tissue will always be left in any FFPE blocks for further diagnostic use. • To prospectively assess the biomarkers identified from previous research please (see table below) by using the Tissue Micro Array and to anticipate material needs for future biological researches: Where tissue permits, all other FFPE sections or blocks will be submitted to the BIOMECA reference laboratory to undergo further processing for biological studies.

Maintaining tissue integrity of the FFPE block for further diagnostic use by the local centre is of paramount importance. Therefore, TMA will only be undertaken if sample size is appropriate. The group recognises that undertaking TMA preparation at an international reference laboratory will optimise preparation of the FFPE resource for all contributing national laboratories and research groups.

DNA will be extracted within the BIOMECA reference laboratory by using standardised robust techniques and quantified. The priority use of the DNA is against the 450k Illumina methylation arrays

However, sufficient tissue will always be left in any FFPE blocks for further diagnostic use, if requested by the local centre. In this case, sufficient remaining FFPE block will be returned to the local investigating centre: • The FFPE blocks will be returned to the local investigating centre within 6 weeks if requested. • The FFPE H&E marked up section will be retained by the BIOMECA reverence laboratory.

From frozen Tissue If multiple pieces of frozen tumour are received, each piece must be individually checked for composition by frozen section and H&E staining, with the help of an experienced pathologist / neuropathologist, to exclude necrotic or normal tissue contamination.

Frozen Tissue Preparation Frozen tissue should be stored at -80ºC or below at all-time pending transfer to the BIOMECA centre Frozen tissue will be used to extract high quality DNA and RNA for high resolution genomic analyses to test the hypothesis that key molecular events are predictive of clinical behaviour of ependymoma. Optional: In countries where touch preparations are prepared for FISH at the local treatment centre, these may be shipped to the national reference pathologist alongside the other required materials, using

Page | 158 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

established national shipping arrangements. At least 10 slides, 5-6 touches each, preferably from two individual tumour pieces, are required. Tumour cell content should be assessed in all tumour pieces used.

The BIOMECA reference laboratories

Biological studies are concentrated in the 4 laboratories from the BIOMECA consortium (Biomarkers for Ependymoma in Children and Adolescents) listed below. Each participating countries (other than France, Germany, Italy and UK) elect to send their samples to one of the 4 primary BIOMECA centres listed below. For example Eire have elected to pair with CBTRC in the UK and Belgium with France. In the chosen primary BIOMECA centre mandatory markers will be performed as well as preparing samples to other BIOMECA centres including DKFZ. Details of samples collected and stored in local centre will be recorded on the eCRF and monitored by BIOMECA consortium. Samples will be transferred from local centre to paired BIOMECA centre on a regular basis as agreed on a country by country basis. This is essential if we are to track where the samples are going, but also to allow the development of trust over sensitive handling of tissue samples and good working practices. The storage, the on-site handling/ manipulation and the cost of transporting samples to BIOMECA centres will be funded by the BIOMECA centres to which samples are sent.

AUSTRIA Dr Christine Haberler Institut of neurology AKH – 4J ; POB48 Währinger Gürtel 18-20 A-1097 Vienna Phone: +43 (0)1-40400-16290 Fax: +43 (0)1-40400-55110 E-mail: [email protected]

FRANCE Dr Jacques Grill CRLCC Institut de cancérologie de l'enfant et de l'Adolescent Institut Gustave Roussy 114 rue Edouard Vaillant 94805 VILLEJUIF CEDEX, Phone: +33 (0)1.42.11.62.09 Fax: +33(0)1.42.11.52.75 E-mail: [email protected]

GERMANY Prof T Pietsch Department of Neuropathology University of Bonn Sigmund-Freud-Str. 25 D-53105 Bonn, Germany Phone: +49 228 287 16602 Fax: +49 228 287 14331 E-mail: [email protected]

Prof S Pfister / Dr H Witt German Cancer Research Center

Page | 159 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Division of Paediatric Neurooncology (B062) Im Neuenheimer Feld 280 D-69120 Heidelberg Germany Phone: +49 6221 424676 E-Mail : [email protected]

ITALY Dr Piergiorgio Modena Laboratory of Cytogenetics and Molecular Genetics Sant' Anna General Hospital Via Ravona, 20 – 22020 Como, Italy Phone: +39 0314859907 /9086 Fax: +39 0314859829 E-mail: [email protected]

UK Professor Richard Grundy/ Dr Lisa Storer Children’s Brain Tumour Research Centre The Medical School (D32) Queens’s Medical Centre Nottingham University, UK Phone: +44 114 8230328 / 8230620 BIOMECA trial phone number: +44 7966 396849 Fax: +44 114 8230696 E-mail: [email protected] Or [email protected]

For further discussion or clarification please contact Professor Richard Grundy Chair BIOMECA consortium

Page | 160 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

EPENDYMOMA TUMOUR SAMPLES

Limited tumour ask neurosurgeon for morer Moderate Plentiful tumour material tumour material tumour material

FFPE: DIAGNOSTIC SAMPLE FFPE RESEARCH FFPE ONLY DIAGNOSTIC FFPE sections DIAGNOSTIC x 1 X1 or BLOCKS (ideally x 2) Central pathological If available, review by national Central pathological Central pathological frozen samples reference pathologist review by national If available, review by national (ideally 6-10 vials) reference pathologist frozen samples reference pathologist

Block will be returned to (Ideally 6-10 vials) local centre Block will be returned to Block will be returned FFPE local centre to local centre sections or Blocks National reference (ideally x4) If consider sufficient available material pathologist for FFPE sections or Blocks may be sent preparation for TMA . National reference pathologist for preparation BIOMECA

BIOMECA Please Note sufficient tissue will always be left in any FFPE blocks for further diagnostic use BIOMECA

Page | 161 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

APPENDIX 6 – VALPROATE PHARMACOKINETICS AND PHARMACODYNAMICS STUDIES (OPTIONAL)

Background Previously published preclinical studies and clinical trials in paediatric patient populations have indicated a clear rationale for the targeting of trough valproate levels, with a view to increasing the likelihood of antitumour activity (Shu et al., 2006; Wolff et al., 2008; Su et al., 2010). However, further work is required to define optimal anticancer valproate concentrations and identify appropriate pharmacodynamics biomarkers of activity and response.

Pharmacokinetic sampling Blood samples (2ml) will be obtained pre-treatment and at 1, 2, 4 and 6h after the initial valproate dose, with an additional sample taken to determine the trough valproate level immediately prior to the next dose or a subsequent dose (for example the following morning dose). This trough level sample is equivalent to a 12h PK sample. Samples will be taken at the same time points following valproate dose escalation where carried out. An additional trough sample will be taken at 12h following valproate administration once the dose resulting in the targeted valproate concentration has been maintained for a period of two weeks (steady state) and at 3 monthly intervals thereafter. All samples will be taken from a central venous line. Blood samples will be immediately transferred to heparinised tubes, centrifuged for 5 min at 2,000 rpm and 4°C and the plasma transferred to a clean labelled tube and stored at -20°C prior to transport to the Northern Institute for Cancer Research (NICR), Newcastle University. A Valproate Pharmacokinetic Sampling Sheet should be completed, with the timing of the samples accurately recorded, and sent to Newcastle along with the samples for analysis. Samples must be sent by overnight courier packed on dry ice in an insulated container. It is essential that either Julie Errington ( [email protected] ; Tel: 0191 208 4357) or Gareth Veal ( [email protected] ; Tel: 0191 208 4332) are contacted prior to sample transport. Mandatory: for centres that are not able to support pharmacokinetic studies we suggest that plasma levels would be measured locally and valproate dosing changed on the basis of local results according valproate dosing table presented in this protocol (see section Valproate ).

Please note that PK sampling times will need to be adjusted accordingly to measure trough valproate levels at an 8h time point for patients who change VPA administration schedule to three times daily dosing.

Dr. Gareth Veal / Julie Errington Northern Institute for Cancer Research Paul O'Gorman Building Medical School Newcastle University Newcastle upon Tyne NE2 4HH UK Tel: +44 (0)191 208 4332 / +44 (0)191 208 4357 Fax: +44 (0)191 208 3452 Email: [email protected] /[email protected]

Page | 162 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

Pharmacokinetic sample analysis An LC/MS assay has been established in Newcastle based on previously published and validated assays for valproate and its major metabolites (Vlase et al., 2008; Gao et al., 2011).

The data obtained from clinical sample analysis will be used to determine valproate pharmacokinetic parameters including area under the plasma concentration-time curve (AUC), clearance and half-life (t½), in addition to trough valproate levels. Trough valproate levels will be used to determine the requirement for dose increments in patients.

Pharmacokinetic modelling will be carried out using these data in conjunction with patient characteristics and clinical parameters in order to investigate the key factors involved in determining individual valproate drug exposures within the patient population.

Pharmacodynamics sampling Blood samples (2ml) will be obtained pre-treatment (prior to the first dose of valproate) and at 4h post valproate administration following the initial valproate dose and at the same time points once the individualised dose resulting in target trough valproate concentrations has been determined. Additional pharmacodynamic samples are to be taken at 6 monthly intervals until the end of treatment. All samples will be taken from a central venous line into Heparin tubes. 1ml of the blood sample will then be immediately transferred to a tube containing fixative (supplied by the NICR, Newcastle University and prepared as detailed in the Lab Manual) and carefully inverted 8 times to ensure mixing of the anticoagulant and preservative. Samples should be maintained at -20°C and shipped to the NICR, Newcastle University on dry ice alongside the PK samples. . It is essential that either Julie Errington ( [email protected]; Tel: 0191 208 4357) or Gareth Veal ( [email protected] ; Tel: 0191 208 4332) are contacted prior to sample transport.

Gareth Veal / Julie Errington Northern Institute for Cancer Research Paul O'Gorman Building Medical School Newcastle University Newcastle upon Tyne NE2 4HH UK Tel: +44 (0)191 208 4332 / +44 (0)191 208 4357 Fax: +44 (0)191 208 3452 Email: [email protected] / [email protected]

Pharmacodynamics sample analysis Pharmacodynamics’ studies will investigate changes in histone H3 and H4 acetylation in fixed circulating leucocytes by immunofluorescent staining of acetylated histones using an ImageStream imaging flow cytometer. Changes in histone H3 and H4 acetylation between baseline and at 4h post-valproate administration at the individualised dose will be correlated with valproate levels and clinical response.

Page | 163 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

APPENDIX 7 – DEFINITION OF HIGHLY EFFECTIVE METHODS OF CONTRACEPTION

A highly effective method of birth control is defined as one which results in a low failure rate (i.e. less than 1% per year) when used consistently and correctly, such as implants, injectables, combined oral contraceptives , some intra uterine devices (IUD), sexual abstinence or vasectomised partner. For subjects using hormonal method, information regarding the products and its potential effects on the contraceptive should be addressed. Acceptable forms of effective contraception should be proposed by the investigator according to the age of the patient, his/her pre or post pubescent status and his/her abstinence. The acceptable forms of effective contraception include:

a) Established use of oral, injected or implanted hormonal methods of contraception. [The decision to allow use of hormonal contraceptives should be based on the Investigational Medicinal Product’s (IMP’s) metabolism and potential for interactions, pharmacology and the adverse event profile (e.g. vomiting)].

b) Placement of an intrauterine device or intrauterine system. [Consideration should be given to the type of device or system being used, as there are higher failure rates quoted for certain types, e.g. steel or copper wire]

c) Barrier methods of contraception: Condom or Occlusive cap (diaphragm or cervical/vault caps) with spermicidal foam/gel/film/cream/pessary. The use of barrier contraceptives should always be supplemented with the use of a spermicide. The following should be noted: - Failure rates indicate that, when used alone, the diaphragm and condom are not highly effective forms of contraception. Therefore the use of additional spermicides does confer additional theoretical contraceptive protection. - However, spermicides alone are inefficient at preventing pregnancy when the whole ejaculate is spilled. Therefore, spermicides are not a barrier method of contraception and should not be used alone.

d) Male sterilisation (with the appropriate post-vasectomy documentation of the absence of sperm in the ejaculate). [For female subjects on the study, the vasectomised male partner should be the sole partner for that subject].

e) True abstinence: When this is in line with the preferred and usual lifestyle of the subject. [Periodic abstinence (e.g., calendar, ovulation, symptothermal, post-ovulation methods) and withdrawal are not acceptable methods of contraception].

Page | 164 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

REFERENCE LIST Reference List

[1] Bouffet E, Capra M, Bartels U. Salvage chemotherapy for metastatic and recurrent ependymoma of childhood. Childs Nerv Syst 2009.

[2] Messahel B, Ashley S, Saran F, et al. Relapsed intracranial ependymoma in children in the UK: patterns of relapse, survival and therapeutic outcome. Eur J Cancer 2009;45:1815-23.

[3] Zacharoulis S, Ashley S, Moreno L, Gentet JC, Massimino M, Frappaz D. Treatment and outcome of children with relapsed ependymoma: a multi-institutional retrospective analysis. Childs Nerv Syst 2009.

[4] Grill J, Le Deley MC, Gambarelli D, et al. Postoperative chemotherapy without irradiation for ependymoma in children under 5 years of age: a multicenter trial of the French Society of Pediatric Oncology. J Clin Oncol 2001;19:1288-96.

[5] Allen JC, Siffert J, Hukin J. Clinical manifestations of childhood ependymoma: a multitude of syndromes. Pediatr Neurosurg 1998;28:49-55.

[6] Bouffet E, Perilongo G, Canete A, Massimino M. Intracranial ependymomas in children: a critical review of prognostic factors and a plea for cooperation. Med Pediatr Oncol 1998;30:319-29.

[7] Chiu JK, Woo SY, Ater J, et al. Intracranial ependymoma in children: analysis of prognostic factors. J Neurooncol 1992;13:283-90.

[8] Conter C, Carrie C, Bernier V, et al. Intracranial ependymomas in children: society of pediatric oncology experience with postoperative hyperfractionated local radiotherapy. Int J Radiat Oncol Biol Phys 2009;74:1536-42.

[9] Duffner PK, Horowitz ME, Krischer JP, et al. The treatment of malignant brain tumors in infants and very young children: an update of the Pediatric Oncology Group experience. Neuro Oncol 1999;1:152-61.

[10] Ernestus RI, Wilcke O, Schroder R. Supratentorial ependymomas in childhood: clinicopathological findings and prognosis. Acta Neurochir (Wien ) 1991;111:96-102.

[11] Grundy RG, Wilne SA, Weston CL, et al. Primary postoperative chemotherapy without radiotherapy for intracranial ependymoma in children: the UKCCSG/SIOP prospective study. Lancet Oncol 2007;8:696-705.

[12] Horn B, Heideman R, Geyer R, et al. A multi-institutional retrospective study of intracranial ependymoma in children: identification of risk factors. J Pediatr Hematol Oncol 1999;21:203-11.

[13] Kuhl J, Muller HL, Berthold F, et al. Preradiation chemotherapy of children and young adults with malignant brain tumors: results of the German pilot trial HIT'88/'89. Klin Padiatr 1998;210:227-33.

[14] Massimino M, Gandola L, Giangaspero F, et al. Hyperfractionated radiotherapy and chemotherapy for childhood ependymoma: final results of the first prospective AIEOP (Associazione Italiana di Ematologia-Oncologia Pediatrica) study. Int J Radiat Oncol Biol Phys 2004;58:1336-45.

Page | 165 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

[15] Massimino M, Gandola L, Barra S, et al. Infant ependymoma in a 10-year AIEOP (Associazione Italiana Ematologia Oncologia Pediatrica) experience with omitted or deferred radiotherapy. Int J Radiat Oncol Biol Phys 2011;80:807-14.

[16] Merchant TE. Current management of childhood ependymoma. Oncology (Williston Park) 2002;16:629-42, 644.

[17] Mork SJ, Loken AC. Ependymoma: a follow-up study of 101 cases. Cancer 1977;40:907-15.

[18] Perilongo G, Massimino M, Sotti G, et al. Analyses of prognostic factors in a retrospective review of 92 children with ependymoma: Italian Pediatric Neuro-oncology Group. Med Pediatr Oncol 1997;29:79-85.

[19] Robertson PL, Zeltzer PM, Boyett JM, et al. Survival and prognostic factors following radiation therapy and chemotherapy for ependymomas in children: a report of the Children's Cancer Group. J Neurosurg 1998;88:695-703.

[20] Kleihues P, Sobin LH. World Health Organization classification of tumors. Cancer 2000;88:2887.

[21] Ellison DW, Kocak M, Figarella-Branger D, et al. Histopathological grading of pediatric ependymoma: reproducibility and clinical relevance in European trial cohorts. J Negat Results Biomed 2011;10:7.

[22] Kilday JP, Mitra B, Domerg C, et al. Copy number gain of 1q25 predicts poor progression-free survival for pediatric intracranial ependymomas and enables patient risk stratification: a prospective European clinical trial cohort analysis on behalf of the Children's Cancer Leukaemia Group (CCLG), Societe Francaise d'Oncologie Pediatrique (SFOP), and International Society for Pediatric Oncology (SIOP). Clin Cancer Res 2012;18:2001-11.

[23] Puget S, Grill J, Valent A, et al. Candidate genes on chromosome 9q33-34 involved in the progression of childhood ependymomas. J Clin Oncol 2009;27:1884-92.

[24] Tabori U, Ma J, Carter M, et al. Human telomere reverse transcriptase expression predicts progression and survival in pediatric intracranial ependymoma. J Clin Oncol 2006;24:1522-8.

[25] Rahman R, Osteso-Ibanez T, Hirst RA, et al. Histone deacetylase inhibition attenuates cell growth with associated telomerase inhibition in high-grade childhood brain tumor cells. Mol Cancer Ther 2010;9:2568-81.

[26] Witt H, Mack SC, Ryzhova M, et al. Delineation of two clinically and molecularly distinct subgroups of posterior fossa ependymoma. Cancer Cell 2011;20:143-57.

[27] Cage TA, Clark AJ, Aranda D, et al. A systematic review of treatment outcomes in pediatric patients with intracranial ependymomas. J Neurosurg Pediatr 2013;11:673-81.

[28] Pejavar S, Polley MY, Rosenberg-Wohl S, et al. Pediatric intracranial ependymoma: the roles of surgery, radiation and chemotherapy. J Neurooncol 2012;106:367-75.

[29] Merchant TE, Li C, Xiong X, Kun LE, Boop FA, Sanford RA. Conformal radiotherapy after surgery for paediatric ependymoma: a prospective study. Lancet Oncol 2009;10:258-66.

[30] Morris EB, Li C, Khan RB, et al. Evolution of neurological impairment in pediatric infratentorial ependymoma patients. J Neurooncol 2009;94:391-8.

Page | 166 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

[31] Grill J, Renaux VK, Bulteau C, et al. Long-term intellectual outcome in children with posterior fossa tumors according to radiation doses and volumes. Int J Radiat Oncol Biol Phys 1999;45:137-45.

[32] Merchant TE. Three-dimensional conformal radiation therapy for ependymoma. Childs Nerv Syst 2009.

[33] Netson KL, Conklin HM, Wu S, Xiong X, Merchant TE. A 5-year investigation of children's adaptive functioning following conformal radiation therapy for localized ependymoma. Int J Radiat Oncol Biol Phys 2012;84:217-23.

[34] Geyer JR, Sposto R, Jennings M, et al. Multiagent chemotherapy and deferred radiotherapy in infants with malignant brain tumors: a report from the Children's Cancer Group. J Clin Oncol 2005;23:7621- 31.

[35] Kilday JP, Mitra B, Domerg C, et al. Copy number gain of 1q25 predicts poor progression-free survival for pediatric intracranial ependymomas and enables patient risk stratification: a prospective European clinical trial cohort analysis on behalf of the Children's Cancer Leukaemia Group (CCLG), Societe Francaise d'Oncologie Pediatrique (SFOP), and International Society for Pediatric Oncology (SIOP). Clin Cancer Res 2012;18:2001-11.

[36] Geyer JR, Sposto R, Jennings M, et al. Multiagent chemotherapy and deferred radiotherapy in infants with malignant brain tumors: a report from the Children's Cancer Group. J Clin Oncol 2005;23:7621- 31.

[37] Zacharoulis S, Levy A, Chi SN, et al. Outcome for young children newly diagnosed with ependymoma, treated with intensive induction chemotherapy followed by myeloablative chemotherapy and autologous stem cell rescue. Pediatr Blood Cancer 2007;49:34-40.

[38] Venkatramani R, Ji L, Lasky J, et al. Outcome of infants and young children with newly diagnosed ependymoma treated on the "Head Start" III prospective clinical trial. J Neurooncol 2013.

[39] White L, Kellie S, Gray E, et al. Postoperative chemotherapy in children less than 4 years of age with malignant brain tumors: promising initial response to a VETOPEC-based regimen. A Study of the Australian and New Zealand Children's Cancer Study Group (ANZCCSG). J Pediatr Hematol Oncol 1998;20:125-30.

[40] Timmermann B, Kortmann RD, Kuhl J, et al. Role of radiotherapy in anaplastic ependymoma in children under age of 3 years: results of the prospective German brain tumor trials HIT-SKK 87 and 92. Radiother Oncol 2005;77:278-85.

[41] Timmermann B, Kortmann RD, Kuhl J, et al. Combined postoperative irradiation and chemotherapy for anaplastic ependymomas in childhood: results of the German prospective trials HIT 88/89 and HIT 91. Int J Radiat Oncol Biol Phys 2000;46:287-95.

[42] Needle MN, Goldwein JW, Grass J, et al. Adjuvant chemotherapy for the treatment of intracranial ependymoma of childhood. Cancer 1997;80:341-7.

[43] Garvin JH, Jr., Selch MT, Holmes E, et al. Phase II study of pre-irradiation chemotherapy for childhood intracranial ependymoma. Children's Cancer Group protocol 9942: a report from the Children's Oncology Group. Pediatr Blood Cancer 2012;59:1183-9.

[44] Kilday JP, Rahman R, Dyer S, et al. Pediatric ependymoma: biological perspectives. Mol Cancer Res 2009;7:765-86.

Page | 167 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

[45] Ridley L, Rahman R, Brundler MA, et al. Multifactorial analysis of predictors of outcome in pediatric intracranial ependymoma. Neuro Oncol 2008;10:675-89.

[46] Witt H, Mack SC, Ryzhova M, et al. Delineation of two clinically and molecularly distinct subgroups of posterior fossa ependymoma. Cancer Cell 2011;20:143-57.

[47] Souweidane MM, Bouffet E, Finlay J. The role of chemotherapy in newly diagnosed ependymoma of childhood. Pediatr Neurosurg 1998;28:273-8.

[48] Needle MN, Goldwein JW, Grass J, et al. Adjuvant chemotherapy for the treatment of intracranial ependymoma of childhood. Cancer 1997;80:341-7.

[49] Souweidane MM, Bouffet E, Finlay J. The role of chemotherapy in newly diagnosed ependymoma of childhood. Pediatr Neurosurg 1998;28:273-8.

[50] Weitman SD, Frazier KM, Kamen BA. The folate receptor in central nervous system malignancies of childhood. J Neurooncol 1994;21:107-12.

[51] Grundy RG, Wilne SA, Weston CL, et al. Primary postoperative chemotherapy without radiotherapy for intracranial ependymoma in children: the UKCCSG/SIOP prospective study. Lancet Oncol 2007;8:696-705.

[52] English MW. 21st century management of ependymoma. Lancet Oncol 2009;10:206-7.

[53] Massimino M, Giangaspero F, Garre ML, et al. Salvage treatment for childhood ependymoma after surgery only: Pitfalls of omitting "at once" adjuvant treatment. Int J Radiat Oncol Biol Phys 2006;65:1440-5.

[54] Aggarwal R, Yeung D, Kumar P, Muhlbauer M, Kun LE. Efficacy and feasibility of stereotactic radiosurgery in the primary management of unfavorable pediatric ependymoma. Radiother Oncol 1997;43:269-73.

[55] Hodgson DC, Goumnerova LC, Loeffler JS, et al. Radiosurgery in the management of pediatric brain tumors. Int J Radiat Oncol Biol Phys 2001;50:929-35.

[56] Stafford SL, Pollock BE, Foote RL, Gorman DA, Nelson DF, Schomberg PJ. Stereotactic radiosurgery for recurrent ependymoma. Cancer 2000;88:870-5.

[57] Grill J, Le Deley MC, Gambarelli D, et al. Postoperative chemotherapy without irradiation for ependymoma in children under 5 years of age: a multicenter trial of the French Society of Pediatric Oncology. J Clin Oncol 2001;19:1288-96.

[58] Duffner PK, Horowitz ME, Krischer JP, et al. Postoperative chemotherapy and delayed radiation in children less than three years of age with malignant brain tumors. N Engl J Med 1993;328:1725-31.

[59] Merchant TE, Mulhern RK, Krasin MJ, et al. Preliminary results from a phase II trial of conformal radiation therapy and evaluation of radiation-related CNS effects for pediatric patients with localized ependymoma. J Clin Oncol 2004;22:3156-62.

[60] Rogers L, Pueschel J, Spetzler R, et al. Is gross-total resection sufficient treatment for posterior fossa ependymomas? J Neurosurg 2005;102:629-36.

Page | 168 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

[61] Geyer JR, Zeltzer PM, Boyett JM, et al. Survival of infants with primitive neuroectodermal tumors or malignant ependymomas of the CNS treated with eight drugs in 1 day: a report from the Childrens Cancer Group. J Clin Oncol 1994;12:1607-15.

[62] Broniscer A, Ke W, Fuller CE, Wu J, Gajjar A, Kun LE. Second neoplasms in pediatric patients with primary central nervous system tumors: the St. Jude Children's Research Hospital experience. Cancer 2004;100:2246-52.

[63] Lannering B, Marky I, Lundberg A, Olsson E. Long-term sequelae after pediatric brain tumors: their effect on disability and quality of life. Med Pediatr Oncol 1990;18:304-10.

[64] Richon VM, O'Brien JP. Histone deacetylase inhibitors: a new class of potential therapeutic agents for cancer treatment. Clin Cancer Res 2002;8:662-4.

[65] Buchwald M, Kramer OH, Heinzel T. HDACi--targets beyond chromatin. Cancer Lett 2009;280:160-7.

[66] Mulhern RK, Merchant TE, Gajjar A, Reddick WE, Kun LE. Late neurocognitive sequelae in survivors of brain tumours in childhood. Lancet Oncol 2004;5:399-408.

[67] Khaw AK, Silasudjana M, Banerjee B, Suzuki M, Baskar R, Hande MP. Inhibition of telomerase activity and human telomerase reverse transcriptase gene expression by histone deacetylase inhibitor in human brain cancer cells. Mutat Res 2007;625:134-44.

[68] Li XN, Shu Q, Su JM, Perlaky L, Blaney SM, Lau CC. Valproic acid induces growth arrest, apoptosis, and senescence in medulloblastomas by increasing histone hyperacetylation and regulating expression of p21Cip1, CDK4, and CMYC. Mol Cancer Ther 2005;4:1912-22.

[69] Rahman R, Osteso-Ibanez T, Hirst RA, et al. Histone deacetylase inhibition attenuates cell growth with associated telomerase inhibition in high-grade childhood brain tumor cells. Mol Cancer Ther 2010;9:2568-81.

[70] Mork SJ, Loken AC. Ependymoma: a follow-up study of 101 cases. Cancer 1977;40:907-15.

[71] Gottlicher M, Minucci S, Zhu P, et al. Valproic acid defines a novel class of HDAC inhibitors inducing differentiation of transformed cells. EMBO J 2001;20:6969-78.

[72] Ecke I, Petry F, Rosenberger A, et al. Antitumor effects of a combined 5-aza-2'deoxycytidine and valproic acid treatment on rhabdomyosarcoma and medulloblastoma in Ptch mutant mice. Cancer Res 2009;69:887-95.

[73] Fouladi M, Furman WL, Chin T, et al. Phase I study of depsipeptide in pediatric patients with refractory solid tumors: a Children's Oncology Group report. J Clin Oncol 2006;24:3678-85.

[74] Atmaca A, Al-Batran SE, Maurer A, et al. Valproic acid (VPA) in patients with refractory advanced cancer: a dose escalating phase I clinical trial. Br J Cancer 2007;97:177-82.

[75] Su JM, Li XN, Thompson P, et al. Phase 1 study of valproic acid in pediatric patients with refractory solid or CNS tumors: a children's oncology group report. Clin Cancer Res 2011;17:589-97.

[76] Wolff JE, Kramm C, Kortmann RD, et al. Valproic acid was well tolerated in heavily pretreated pediatric patients with high-grade glioma. J Neurooncol 2008;90:309-14.

Page | 169 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

[77] Conklin HM, Li C, Xiong X, Ogg RJ, Merchant TE. Predicting change in academic abilities after conformal radiation therapy for localized ependymoma. J Clin Oncol 2008;26:3965-70.

[78] Landau E, Boop FA, Conklin HM, Wu S, Xiong X, Merchant TE. Supratentorial ependymoma: disease control, complications, and functional outcomes after irradiation. Int J Radiat Oncol Biol Phys 2013;85:e193-e199.

[79] Di PM, Conklin HM, Li C, Xiong X, Merchant TE. Investigating verbal and visual auditory learning after conformal radiation therapy for childhood ependymoma. Int J Radiat Oncol Biol Phys 2010;77:1002- 8.

[80] Charalambides C, Dinopoulos A, Sgouros S. Neuropsychological sequelae and quality of life following treatment of posterior fossa ependymomas in children. Childs Nerv Syst 2009;25:1313-20.

[81] von HK, Kieffer V, Habrand JL, Kalifa C, Dellatolas G, Grill J. Impairment of intellectual functions after surgery and posterior fossa irradiation in children with ependymoma is related to age and neurologic complications. BMC Cancer 2008;8:15.

[82] Moreno L, Pollack IF, Duffner PK, et al. Utility of cerebrospinal fluid cytology in newly diagnosed childhood ependymoma. J Pediatr Hematol Oncol 2010;32:515-8.

[83] Zacharoulis S, Ji L, Pollack IF, et al. Metastatic ependymoma: a multi-institutional retrospective analysis of prognostic factors. Pediatr Blood Cancer 2008;50:231-5.

[84] Payet M, Conter C, Labrousse F, et al. Clear cell ependymoma with trisomy 19 developing bone metastases. Childs Nerv Syst 2012;28:739-42.

[85] Bouffet E, Hawkins CE, Ballourah W, et al. Survival benefit for pediatric patients with recurrent ependymoma treated with reirradiation. Int J Radiat Oncol Biol Phys 2012;83:1541-8.

[86] Bouffet E, Capra M, Bartels U. Salvage chemotherapy for metastatic and recurrent ependymoma of childhood. Childs Nerv Syst 2009;25:1293-301.

[87] Schellinger KA, Propp JM, Villano JL, McCarthy BJ. Descriptive epidemiology of primary spinal cord tumors. J Neurooncol 2008;87:173-9.

[88] Benesch M, Weber-Mzell D, Gerber NU, et al. Ependymoma of the spinal cord in children and adolescents: a retrospective series from the HIT database. J Neurosurg Pediatr 2010;6:137-44.

[89] Feeny D, Furlong W, Boyle M, Torrance GW. Multi-attribute health status classification systems. Health Utilities Index. Pharmacoeconomics 1995;7:490-502.

[90] Kennedy CR, Leyland K. Comparison of screening instruments for disability and emotional/behavioral disorders with a generic measure of health-related quality of life in survivors of childhood brain tumors. Int J Cancer Suppl 1999;12:106-11.

[91] Varni JW, Seid M, Rode CA. The PedsQL: measurement model for the pediatric quality of life inventory. Med Care 1999;37:126-39.

[92] Varni JW, Burwinkle TM, Katz ER, Meeske K, Dickinson P. The PedsQL in pediatric cancer: reliability and validity of the Pediatric Quality of Life Inventory Generic Core Scales, Multidimensional Fatigue Scale, and Cancer Module. Cancer 2002;94:2090-106.

Page | 170 CONFIDENTIAL Ependymoma Program II FINAL Protocol Version 3. 1 April 22 th , 2020

[93] Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst 1993;85:365-76.

[94] Smets EM, Garssen B, Bonke B, de Haes JC. The Multidimensional Fatigue Inventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psychosom Res 1995;39:315-25.

[95] Glaser A, Kennedy C, Punt J, Walker D. Standardized quantitative assessment of brain tumor survivors treated within clinical trials in childhood. Int J Cancer Suppl 1999;12:77-82.

[96] Ivanova A, Qaqish BF, Schell MJ. Continuous toxicity monitoring in phase II trials in oncology. Biometrics 2005;61:540-5.

[97] Massimino M, Miceli R, Giangaspero F et al. Final results of the second prospective AIEOP protocol for pediatric intracranial ependymoma. Neuro Oncol. 2016 Oct;18(10):1451-60.

[98] Grundy RG, Wilne SH, Robinson KJ et al. Primary postoperative chemotherapy without radiotherapy for treatment of brain tumours other than ependymoma in children under 3 years: results of the first UKCCSG/SIOP CNS 9204 trial. Eur J Cancer. 2010 Jan;46(1):120-33. doi: 10.1016/j.ejca.2009.09.013.

Page | 171 CONFIDENTIAL