ORIGINAL ARTICLE. Children s Hospital Augsburg, Augsburg, Germany; Germany;

Transcription

1 Bone Marrow Transplantation (2014) 49, & 2014 Macmillan Publishers Limited All rights reserved /14 ORIGINAL ARTICLE High-dose chemotherapy (HDCT) with auto-sct in children with atypical teratoid/rhabdoid tumors (AT/RT): a report from the European Rhabdoid Registry (EU-RHAB) M Benesch 1, K Bartelheim 2, G Fleischhack 3, B Gruhn 4, PG Schlegel 5, O Witt 6, KD Stachel 7, H Hauch 8, C Urban 1, F Quehenberger 9, M Massimino 10, T Pietsch 11, M Hasselblatt 12, F Giangaspero 13,14, U Kordes 15, R Schneppenheim 15, P Hauser 16, T Klingebiel 17 and MC Frühwald 2,18 A retrospective analysis of data from the European Rhabdoid Registry (EU-RHAB) was performed to describe the outcome of children with atypical teratoid/rhabdoid tumors (AT/RT) who underwent high-dose chemotherapy (HDCT) with auto-sct. Nineteen patients (male, n ¼ 15; median age at diagnosis 21 months) were identified. Nine patients presented with metastatic disease at diagnosis. A partial or subtotal resection was achieved in 11, a total resection in five and a biopsy in three patients. Patients received a median of six chemotherapy cycles prior to HDCT. Additional radiotherapy was performed in 14 patients (first-line, n ¼ 9; following progression, n ¼ 5). Six patients underwent tandem auto-sct. Disease status before HDCT was CR in six, PR in eight, stable disease in two and progressive disease (PD) in two patients (data missing, n ¼ 1). With a median follow-up of 16 months, 14 patients progressed. Estimated progression-free and OS at 2 years were 29% (±11%) and 50% (±12%), respectively. At last follow-up, eight patients were alive (first CR, n ¼ 4; second CR, n ¼ 2; PR, n ¼ 1; PD, n ¼ 1). Eleven patients died of PD. Median time-to-progression was 14 months. Selected patients with AT/RT might benefit from HDCT with radiotherapy. The definitive impact of this treatment modality has to be evaluated prospectively in a randomized trial. Bone Marrow Transplantation (2014) 49, ; doi: /bmt ; published online 13 January 2014 Keywords: atypical teratoid/rhabdoid tumors; auto-sct; children INTRODUCTION Atypical teratoid/rhabdoid tumors (AT/RT) are highly malignant tumors of the central nervous system (CNS). 1 Among children below the age of 3 years, AT/RT constitute the most common malignant CNS tumors (17.3%), followed by medulloblastomas (16%) and primitive neuroectodermal tumors of the CNS (13.3%). 2 In a population-based study, the age-standardized incidence rate of AT/RT was calculated at 1.38 per person years. 2 The median age at diagnosis is between 14 and 24 months, with boys more commonly affected than girls. 1 9 Infratentorial location has been reported in 38 65% of AT/RT and up to 1/3 of patients with AT/RT present with metastatic disease at diagnosis. 1 9 Prognosis of patients with AT/RT is generally poor. The vast majority of patients develop early local progression or recurrence±distant metastases within 1 year following diagnosis. 1 8 Among various clinical parameters, younger age and presence of metastases were most consistently associated with a worse outcome. 4,5 7,9 In terms of treatment, a gross total resection 2,4,5,8 and radiotherapy 5,6,9 seem to contribute to a prolongation of survival. The impact of high-dose chemotherapy (HDCT) with auto-sct on prognosis is unclear. The aim of the present retrospective analysis was thus to report the clinical characteristics and outcome of 19 children with AT/RT who underwent HDCT within the European Rhabdoid Registry (EU-RHAB) and the previous registry Rhabdoid PATIENTS AND METHODS Aims of EU-RHAB and rhabdoid 2007 The primary goal of the EU-RHAB registry is to optimize the management of patients with rhabdoid tumors of all anatomical sites. Thus EU-RHAB serves as a platform for the standardized registration of epidemiologic-, molecular-, clinical- and treatment-related data. EU-RHAB contains 1 Division of Pediatric Hematology and Oncology, Department of Pediatrics and Adolescent Medicine, Medical University of Graz, Graz, Austria; 2 Swabian Children s Cancer Center, Children s Hospital Augsburg, Augsburg, Germany; 3 Pediatrics III, University Hospital of Essen, Essen, Germany; 4 Department of Pediatrics, Jena University Hospital, Jena, Germany; 5 Department of Pediatric Hematology, Oncology and Neurooncology, University Children s Hospital Würzburg, Würzburg, Germany; 6 Department of Pediatric Oncology, Hematology, Immunology and Pneumonology, Children s Hospital, University of Heidelberg, Heidelberg, Germany; 7 Children s University Hospital, Friedrich-Alexander University Erlangen-Nürnberg, Erlangen, Germany; 8 Department of Pediatric Hematology and Oncology, Justus Liebig University Gieen, Gieen, Germany; 9 Institute for Medical Statistics, Medical University of Graz, Graz, Austria; 10 Department of Pediatrics, Fondazione IRCCS, Istituto Nazionale Tumori, Milan, Italy; 11 Institute of Neuropathology, University of Bonn, Bonn, Germany; 12 Institute of Neuropathology, University Hospital Münster, Münster, Germany; 13 Department of Radiological, Oncological and Anatomo-pathological Sciences, University La Sapienza, Rome, Italy; 14 IRCCS Neuromed, Pozzilli, Italy; 15 Department of Pediatric Hematology and Oncology, University Medical Center Hamburg- Eppendorf, Hamburg, Germany; 16 Department of Pediatrics, Semmelweis University, Budapest, Hungary; 17 Division of Pediatric Hematology and Oncology, Department of Pediatrics, J. W. Goethe University Children s Hospital of Frankfurt, Frankfurt, Germany and 18 Department of Pediatric Hematology and Oncology, University Children s Hospital Münster, Münster, Germany. Correspondence: Dr M Benesch, Division of Pediatric Hematology and Oncology, Department of Pediatrics and Adolescent Medicine, Medical University of Graz, Auenbruggerplatz 38, A-8036 Graz, Austria or Dr MC Frühwald, Swabian Children s Cancer Center, Children s Hospital Augsburg, Stenglinstrae 2, D Augsburg, Germany. martin.benesch@klinikum-graz.at or michael.fruehwald@klinikum-augsburg.de Received 9 July 2013; revised 12 November 2013; accepted 12 November 2013; published online 13 January 2014

2 treatment recommendations based on the best available evidence including published data sets, the investigators own experiences and data from the German, Austrian and Swiss Society of Pediatric Oncology and Hematology and the Société International d Oncologie Pédiatrique. These recommendations represent a non-investigational consensus standard of care derived from currently available data. The previous Rhabdoid 2007 registry had the same objectives; however, treatment recommendations were less detailed compared with those of the EU-RHAB. Also, it did not reach an uniform European consensus, as it did not include actinomycin D and contained VCR in each element of chemotherapy. Between October 2005 and August 2011, six patients were enrolled in the Rhabdoid 2007 and 13 in the subsequent EU-RHAB registry. It was prospectively anticipated to officially initiate the Rhabdoid registry in However, data collection started earlier in 2005 explaining the fact that four patients were treated before the formal initiation of the Rhabdoid 2007 registry. EU-RHAB has been approved by the ethics committee of the University of Münster, Germany. Informed consent was obtained from all patients by parents or guardians prior to the initiation of therapy. Diagnostic evaluation Complete CNS staging including contrast-enhanced magnetic resonance imaging (MRI) of the entire neuraxis and cytological analysis of cerebrospinal fluid (CSF) was strongly recommended for all patients at diagnosis. Extraneural metastases were excluded by clinical examination and imaging studies. Follow-up MRI was recommended following cycles 2, 4, 6 and 9 of intensive chemotherapy. However, the clinical course often necessitated MRI at shorter intervals. Complete neuraxis staging was performed in 16/19 patients (84%). Central pathological review was performed at the time of diagnosis in 18 patients (95%) either at the Neuropathology Reference Centre of the HIT network (Institute of Neuropathology, University of Bonn, Germany, n ¼ 12), the Institute of Neuropathology, University Hospital Münster, Germany (n ¼ 4), or at the Department of Pathology, University La Sapienza, Rome, Italy (n ¼ 2). In one patient, diagnosis was established only by cytology of CSF cells followed by FISH for SMARCB1 deletion as the patient was deemed inoperable. Genetic testing for germline mutations of SMARCB1 was performed in 13 patients (68%) as previously described; 10 in four of them (patients 5, 6, 13 and 15) a germline mutation was identified. Chemotherapy (Rhabdoid 2007) Postoperative chemotherapy contained a total of nine cycles of alternating VCD and ICE (see below) with concomitant age-adjusted intraventricular MTX via the Ommaya or Rickham reservoir (Supplementary Figure 1). The registry documents of Rhabdoid 2007 and EU-RHAB provided also detailed guidelines on the administration of intraventricular MTX in the presence of a ventriculoperitoneal (VP) shunt. In our series two patients had a VP shunt with a pressure regulating valve and 2 subduroperitoneal shunts. In all cases the guidelines were strictly followed. VCD consisted of VCR (1.5 mg/m 2 /day; days 1, 8 and 15), CY (1800 mg/ m 2 /day; day 1) and doxorubicin (37.5 mg/m 2 /day; days 1 and 2). ICE consisted of ifosfamide (2 g/m 2 /day; days 1 3), carboplatin (600 mg/m 2 / day; day 1) and etoposide (100 mg/m 2 /day; days 1 3). Drug doses were calculated in kg body weight in children under the age of 12 months (VCR o36 months). treatment. In addition, patients who did not receive chemotherapy at full doses were candidates for maintenance therapy. Radiotherapy In children X18 months or older, radiotherapy had to be started as soon as possible. Children o18 months were irradiated only under exceptional circumstances. In the case of primary metastatic disease, RT might be delayed until the end of intensive chemotherapy (following cycle 9). Postponing radiotherapy was recommended in children p18 months until age permitted its use or following cycle 9. In children X18 months with localized AT/RT, the extended tumor region was irradiated with 54.0 Gy (1.8 Gy/day, 5 fractions/week). Patients with metastatic disease (Chang M1 M3, age 418 months to 3 years) were scheduled to receive craniospinal radiotherapy with 24.0 Gy (1.6 Gy/day). Boosts were delivered to the primary tumor site (54 Gy, 1.8 Gy/day) and circumscribed solid spinal and intracranial metastases (49.2 Gy, 1.8 Gy/day). In children older than 3 years with metastatic disease (Chang M1 M3), the craniospinal axis dose was 35.2 Gy. High-dose chemotherapy Within the Rhabdoid 2007/EU-RHAB registries, the use of HDCT was at the discretion of the treating physician (Supplementary Figures 2 and 3). As individual centers and some countries prefer to employ HDCT in patients with AT/RT, the protocol aimed to harmonize transplant strategies. The HDCT protocol recommended a conditioning regimen based on carboplatin (500 mg/m 2 /day; days 6to 4) and thiotepa (300 mg/m 2 /day; days 6to 4). Stem cell harvest had to be performed after the first ICE cycle. If indicated, stem cell collection might be repeated after the second ICE cycle. Stem cell mobilization and processing were performed according to standard procedures. A minimum of CD34 þ stem cells per kg body weight was required for transplantation. Other conditioning regimens (Table 2) used were carboplatin (500 mg/m 2 /day on days 8 to 5) and etoposide (250 mg/m 2 /day on days 8to 5) for the first and thiotepa (300 mg/m 2 /day on days 4to 2) and CY (1500 mg/m 2 / day on days 4to 2) for the second transplantation in five patients who underwent tandem auto-sct. In one patient who received tandem auto-sct conditioning consisted of thiotepa (150 mg/m 2 /day on days 5 to 4) and melphalan (40 mg/m 2 /day on days 5to 3 and 20 mg/m 2 / day on day 2 (total melphalan dose 140 mg/m 2 )) for the first and second transplant. Three patients received conditioning with carboplatin and etoposide (doses as above). Data extraction and statistical analysis All data were extracted without direct personal identification. Treatment centers were contacted by the EU-RHAB study center in order to obtain the most recent follow-up information. Events other than relapse or progression did not occur. The influence of risk factors (age, gender, metastases at diagnosis, extent of resection, achievement of CR prior to HDCT, radiotherapy and presence of SMARCB1 germline mutations) in disease progression was tested using the log-rank test. R (www. r-project.org) and package survival version were used for calculations. Asymptotic 95% confidence limits were calculated on the log log of the survival function. Results were updated as of 31 January Chemotherapy (EU-RHAB) Chemotherapy also included nine cycles of doxorubicin, ICE and VCA with concomitant age-adjusted intraventricular MTX (o2 years: 0.5 mg, 2 3 years: 1 mg, 3 years: 2 mg; 4 doses during doxorubicin, 3 doses during VCA and 4 doses during ICE cycles) (Supplementary Figures 2 and 3). Doxorubicin was given as a single agent at 37.5 mg/m 2 /day on days 1 and 2. ICE chemotherapy was identical to that used in Rhabdoid 2007 except for a slight dose reduction of carboplatin (500 mg/m 2 /day). VCA consisted of VCR (1.5 mg/m 2 /day; days 1 and 8), CY (1500 mg/m 2 /day; day 1) and actinomycin D (25 mg/m 2 /day; days 1 and 2). For infants o6 months or o10 kg, doses were further reduced to 2/3 or 1/2 in initial cycles and were adjusted according to tolerance. Maintenance chemotherapy Maintenance therapy with trofosfamide, idarubicine, etoposide and temozolomide was recommended in patients with germline mutations, residual disease after termination of intensive therapy or slow response to RESULTS Clinical characteristics Basic clinical characteristics of study patients are shown in Table 1. The median age at diagnosis was 21 months (range, 0 64) with a strong male predominance of 3.75:1. Nine patients (47%) had metastatic disease at diagnosis (positive CSF, n ¼ 4; solid CNS metastases, n ¼ 2; positive CSF and solid CNS metastases, n ¼ 3). Tumor location was supratentorial in 11 and infratentorial in eight patients. Two of the patients with SMARCB1 germline mutations presented with synchronous peripheral malignant rhabdoid tumors (Table 1). One patient (patient 13) has been reported previously. 11 Briefly, a 2-year-old boy with mosaic Klinefelter syndrome was primarily found to have metastatic medulloblastoma and was treated according to the Italian protocol for high-risk medulloblastoma. He was in CR until the age of 7½ years when he developed an isolated spinal lesion, which was subtotally & 2014 Macmillan Publishers Limited Bone Marrow Transplantation (2014)

3 372 Table 1. Basic clinical characteristics of study patients Patient Gender Age at diagnosis (mos) Tumor location Dissemination at diagnosis Initial surgery 1 Male 31 Supratentorial No a Partial resection 2 Male 22 Supratentorial Yes (CSF) Subtotal resection 3 Male 11 Infratentorial No Partial resection 4 Female 11 Infratentorial No Total resection 5 Female 0 Supratentorial, synchronous No Total resection temporo-/ periorbital MRT 6 Male 4 Infratentorial/ central midline, Yes (CSF þ solid) Total resection synchronous cervical MRT 7 Female 42 Supratentorial Yes (CSF þ solid) Biopsy 8 Male 10 Infratentorial No b Subtotal resection 9 Male 64 Infratentorial No Partial resection 10 Male 1 Infratentorial No Biopsy 11 Male 35 Supratentorial Yes (CSF) Partial resection 12 Male 6 Supratentorial Yes (solid) Subtotal resection 13 c Male 24 Infratentorial Yes (CSF þ solid) Partial resection 14 Male 52 Supratentorial No Partial resection 15 Male 29 Supratentorial Yes (CSF) b Total resection 16 Male 29 Supratentorial Yes (solid) Biopsy 17 Male 21 Supratentorial Yes (CSF) Partial resection 18 Female 10 Supratentorial No Total resection 19 Male 13 Infratentorial No Partial resection Abbreviations: Mos ¼ months; MRT ¼ malignant rhabdoid tumor. a No cytological analysis of CSF. b No spinal MRI. c Spinal relapse of an infratentorial tumor originally considered to be medulloblastoma was treated according to the EU-RHAB. Table 2. Treatment characteristics of study patients Patient Number of chemotherapy cycles prior to HDCT Response to induction chemotherapy Radiotherapy Conditioning Disease status after HDCT Clinical course and follow-up (mos) 1 6 SD No CBCDA/ETO PD DOD (8) 2 6 n.a. No First auto-sct: CBCDA/ETO CR PD-DOD (11) 3 8 CR Yes (50 Gy, following CBCDA/ETO PD PD-DOD (14) progression þ SL surgery) 4 9 PR Yes (54 Gy, following progression) CBCDA/TTP CR PD-CR2/NED (48) 5 8 CR No First auto-sct: CBCDA/ETO CR NED (49) 6 6 PR Yes (54 Gy, following progression) CBCDA/TTP PR PD-DOD (27) 7 6 PD Yes (30 Gy, CSI (early CBCDA/TTP SD PD-DOD (11) discontinuation), following progression) 8 6 CR Yes (54 Gy, prior to progression) CBCDA/TTP CR PD-AWD (34) 9 6 CR Yes (54 Gy, prior to progression) CBCDA/TTP PD PD-DOD (15) 10 5 PR No First auto-sct: CBCDA/ETO SD PD-DOD (12) 11 7 SD Yes (50 Gy, prior to progression) CBCDA/TTP PD PD-DOD (16) 12 2 PD Yes (20 Gy, area of recurrence First/second auto-sct: SD PD-DOD (10) only) TTP/MEL 13 a 6 PR Yes (40 Gy) CBCDA/TTP PR PD-CR2 (91) 14 6 CR Yes (54 Gy, prior to progression) CBCDA/TTP CR PD-DOD (22) 15 3 PR No CBCDA/ETO CR PD-DOD (18) 16 6 PR Yes (55 Gy, CSI (35.2 Gy)) CBCDA/TTP PR PR-AWD (16) 17 8 PR Yes (60 Gy) CBCDA/TTP PR CR-NED (15) 18 4 CR Yes (54 Gy) First auto-sct: CBCDA/ETO CR CR-NED (83) 19 6 PR Yes (54 Gy, 6 Gy boost) First auto-sct: CBCDA/ETO PR CR-NED (43) Abbreviations: AWD ¼ alive with disease; CBCDA ¼ carboplatin; CPM ¼ CY; CSI ¼ craniospinal irradiation; DOD ¼ dead of disease; ETO ¼ etoposide; MEL ¼ melphalan; n.a. ¼ not available; NED ¼ no evidence of disease; PD ¼ progressive disease; SD ¼ stable disease; SL ¼ second look; TTP ¼ thiotepa. a Spinal relapse of an infratentorial tumor originally considered to be medulloblastoma was treated according to the EU-RHAB. Bone Marrow Transplantation (2014) & 2014 Macmillan Publishers Limited

4 removed. Histopathological (re-) evaluation of both the spinal and previous cerebellar lesion revealed AT/RT. Genetic testing for rhabdoid tumor predisposition syndrome was positive. Subsequent treatment was performed according to the EU-RHAB recommendations. Treatment prior to high-dose chemotherapy Treatment-related data of study patients are summarized in Tables 1 and 2. All patients underwent initial surgery (total resection, n ¼ 5; partial or subtotal resection, n ¼ 11; biopsy, n ¼ 3) and conventional induction chemotherapy prior to HDCT. The median number of pretransplant chemotherapy cycles was 6 (range, 2 9). As the Rhabdoid 2007 registry did not contain any specific recommendation concerning HDCT, it was at the discretion of the treating physician when to proceed with HDCT. Ten of the 13 patients included in the EU-RHAB completed the six planned chemotherapy courses prior to HDCT as per protocol. Two patients received one and two additional courses at the discretion of the treating physicians, respectively. The remaining patient progressed under chemotherapy and received only 5 cycles. Radiotherapy was performed in 14 patients (first-line, n ¼ 9; following progression, n ¼ 5); 12 of these patients received local and two received craniospinal radiotherapy. High-dose chemotherapy The median age at the time of transplantation was 29 months (range, 6 98). HDCT was used as part of the first-line treatment in 12 study patients. In seven patients, HDCT was performed as salvage treatment following first relapse or progression. Six study patients underwent tandem auto-sct. Carboplatin/thiotepa (n ¼ 10) was the most commonly employed regimen, followed by carboplatin/etoposide (n ¼ 8), CY/thiotepa (n ¼ 5) and thiotepa/ melphalan regimens (n ¼ 2) (Table 2). In the vast majority of patients, conditioning was based on carboplatin (n ¼ 18) and/or thiotepa (n ¼ 17). Disease status before HDCT was CR in six, PR in eight, stable disease in two and progressive disease (PD) in two patients (data missing, n ¼ 1). In two patients, remission status (CR/PR) was achieved by second look surgery following relapse. A median of CD34 þ cells per kg body weight (range, ) was infused. All patients engrafted. TRM was 0%. Clinical course and status of remission With a median follow-up of 16 months (range, 8 83), 14 patients (74%) progressed. Estimated PFS and OS at 2 years were 29% (±11%) and 50% (±12%), respectively (Figures 1a and b). At last follow-up, eight patients were alive (first CR, n ¼ 4; second CR, n ¼ 2; PR, n ¼ 1; PD, n ¼ 1). Eleven patients died of PD. The median estimated time-to-progression was 14 months. The first recurrence was a local progression in five, a locoregional relapse in four and a distant relapse in four patients. One patient had a combined local and distant progression. None of the parameters tested had an impact on PFS or OS. Of seven patients who underwent HDCT following progression or relapse, two (patients 4 and 13) were alive at last follow-up in second CR 32 and 24 months after relapse (28 and 17 months after HDCT). Both patients had received additional local radiotherapy following relapse. Of 12 patients who were transplanted as part of their first-line treatment, four (patients 5, 17, 18 and 19) were in first continuous CR 49, 15, 83 and 43 months after diagnosis. Ages at diagnosis of these patients were 0, 21, 20 and 13 months. Three had a supratentorial and one an infratentorial AT/RT. One patient (patient 17) had positive CSF cytology at the time of diagnosis; none had solid metastases. Local radiotherapy at Gy was delivered in three patients (prior to HDCT, n ¼ 1; following HDCT, n ¼ 2). a Progression-free survival b Overall survival Patients Months Months Figure 1. (a) Kaplan Meier plots of the estimated PFS rate for all 19 study patients. (b) Kaplan Meier plots of the estimated OS rate for all 19 study patients. Of note, two patients with germline SMARCB1 mutations are in first (patient 5) or second (patient 13) CR. The two other patients with germline SMARCB1 mutations died of PD 26 and 18 months after diagnosis. Late effects in long-term survivors The late effects of the previously reported patient are described in detail elsewhere. 11 Of the remaining four long-term survivors (X24 months follow-up from diagnosis), one (patient 19) suffers from eye movement dysfunction (strabismus), mild ataxia and mild dyslalia. One patient (patient 18) has tumor-associated leftsided hemiparesis and facial nerve paresis already present at the time of diagnosis. Both patients developed bilateral sensorineural hearing loss requiring a hearing aid in one of them and (currently unsubstituted) growth hormone deficiency. They also demonstrate delay of language development; hence they attend a special preschool class and receive additional speech and educational training. A developmental delay, ataxia of the trunk, amblyopia and strabism were observed in patient 4, whereas in patient 5, who attends a normal kindergarten, no late sequelae are documented. DISCUSSION Although intensive multimodality treatment 12,13 has been shown to improve survival in a proportion of patients with AT/RT, data concerning the use of HDCT are still conflicting and limited to case reports and few retrospective case series. 5,8,14 21 These reports are 373 & 2014 Macmillan Publishers Limited Bone Marrow Transplantation (2014)

5 374 Table 3. Reports on HDCT and auto-sct in children with AT/RT Reference Number of patients (M status) Median age (mos) Surgery Radiotherapy Status prior to HDCT Conditioning Outcome (survival (mos)) 5 n ¼ 13 (n.r.) 31 GTR, n ¼ 7 STR, n ¼ 6 n ¼ 5 (CSI, n ¼ 2) NED, n ¼ 9 MD, n ¼ 4 CBCDA/TTP, n ¼ 7 Other regimens, n ¼ 6 NED, n ¼ 6 (9.5 90) DOD, n ¼ 6, DOC, n ¼ 1 8 a,b n ¼ 9(Mþ, n ¼ 3) 19.8 GTR, n ¼ 5 ogtr, n ¼ 4 n ¼ 3 (CSI, n ¼ 1) n.r. CBCDA/TTP/ETO, n ¼ 1 CBCDA/TTP ( 3), n ¼ 8 NED, n ¼ 8 ( ) 14 n ¼ 1(M0) 4 ogtr MD CBCDA/TTP/ETO (First HDCT) NED (32) BU/MEL/TTP (second HDCT) 15 n ¼ 3 (M0) 29, 42, 91 GTR, n ¼ 2 ogtr, n ¼ 1 n ¼ 3 NED, n ¼ 3 ETO/TTP/CPM NED, n ¼ 2(101þ /105 þ ) 16 n ¼ 6(Mþ, n ¼ 2) 24 GTR, n ¼ 5 ogtr, n ¼ 1 n ¼ 1 NED, n ¼ 6 Various regimens NED, n ¼ 2 (78/98) DOD, n ¼ 3 17 b n ¼ 6(Mþ, n ¼ 3) 35 GTR, n ¼ 2 n ¼ 2 n.r. CBCDA/TTP (x3) NED, n ¼ 4 (23 64) STR, n ¼ 4 18 c n ¼ 6(Mþ, n ¼ 4) 11.5 GTR, n ¼ 4 ogtr, n ¼ 2 19 c n ¼ 5(Mþ, n ¼ 5) 14.5 GTR, n ¼ 3 ogtr, n ¼ 2 20 d n ¼ GTR, n ¼ 7 ogtr, n ¼ 6 21 d n ¼ 32 HSI: 36 HSII: 28 HSIII: 13.6 GTR/near GTR HSI: 33% HSII: 72% HSIII: 73% n ¼ 5 NED, n ¼ 2 MD, n ¼ 4 CBCDA/TTP/ETO (First HDCT) CPM/MEL (second HDCT) n ¼ 2 MD, n ¼ 5 CBCDA/TTP/ETO (First HDCT) n ¼ 4 (CSI, n ¼ 2) NED, n ¼ 6 MD, n ¼ 5 n.r., n ¼ 2 DOD, n ¼ 2 NED, n ¼ 4 (16 70) DOD, n ¼ 1 PD, n ¼ 5 CPM/MEL (second HDCT) CBCDA/TTP/ETO NED, n ¼ 3 (42 þ,54þ,67þ) DOD, n ¼ 8 DOC, n ¼ 2 n.r. CBCDA/TTP/ETO 2-year EFS (HSI): n.r. 2-year EFS (HSII): 43% 2-year EFS (HSIII): 28% Abbreviations: AT/RT ¼atypical teratoid/rhabdoid tumors; AWD ¼ alive with disease; CSI ¼ craniospinal irradiation; DOC ¼ dead of other causes; DOD ¼ dead of disease; GTR ¼ gross total resection; ogtr ¼ less than gross total resection; HDCT ¼ high-dose chemotherapy; HS ¼ Head Start; M0 ¼ no metastases at the time of diagnosis; M þ¼evidence of metastases at the time of diagnosis; MD ¼ measurable disease; n.r. ¼ not reported; NED ¼ no evidence of disease; PD ¼ progressive disease; STR ¼ subtotal resection. a A total of 19 patients underwent HDCT with auto-sct. Detailed data of the nine survivors were provided in the paper. b,c,d Overlapping study populations. summarized in Table 3. The number of patients included in these studies ranges from 3 to 32. 5,8,14 21 In the larger series, a 2-year (event-free or overall) survival of up to 50% has been reported. 8,21 Recently, a retrospective, non-randomized Canadian study has demonstrated a statistically significant survival benefit in patients who had undergone HDCT. 8 A descriptive analysis of 26 long-term survivors after HDCT arbitrarily defined as surviving without evidence of disease X24 months from diagnosis included in the present and previous series 5,8,14 18,20 revealed a median age at diagnosis of 18 months (range, 0 49). Supra- and infratentorial AT/RT were almost equally represented (12:14). In the vast majority of patients (n ¼ 19), the tumor was localized at diagnosis; only three patients had metastatic disease (data on the disease stage not available, n ¼ 4). A complete resection was achieved in 17 patients. Radiotherapy was administered in seven patients as part of initial treatment and in five patients at relapse. Fourteen patients (53%) did not receive any radiotherapy. Of note, a small number of patients (n ¼ 6) in these studies were in remission but had shorter (o24 months) follow-up. The high proportion of non-irradiated patients indicates that HDCT is used in a considerable number of patients to avoid radiotherapy and, in fact, this approach might occasionally be successful. Another goal of HDCT is to rescue patients who had experienced disease progression. Five of the 26 patients who survived 42 years were transplanted following PD and achieved a second remission. All five patients underwent irradiation at the time of relapse; in three of them second surgery was performed. Thus, radiotherapy seems to be critically important in achieving a second remission following PD. Three of the 26 patients had metastases at diagnosis, indicating that only a small minority of patients with disseminated disease will ultimately reach long-term survival. So far, HDCT has generally been well tolerated with low TRM. One out of 13 patients died from toxic death in the study by Hilden et al. 5 Among patients (n ¼ 7) treated according to the Head Start II protocol, one patient died from meningitis and another from secondary leukemia. 20 In other studies, no treatment-related deaths were explicitly reported. 8,15 18 In contrast, the best treatment results in patients with AT/RT reported to date were achieved by an intensive multimodality treatment based on systemic and intrathecal chemotherapy and radiotherapy. PFS and OS at 2 years were 53% and 70%, respectively. Craniospinal and primary site doses in the Rhabdoid 2007 and EU-RHAB registries were identical to those used in the study by Chi et al. 12 which enrolled 20 patients with AT/RT. Eleven of them received conformal RT and two of them relapsed. Among the four patients who received craniospinal irradiation, three have experienced relapse. Age was a significant prognostic parameter in some, 4,5,7,9 but not all studies. 6,8,12 As the median age at diagnosis of patients who receive radiotherapy is higher than in non-irradiated patients, 4,5 the better outcome of older patients most probably reflects the fact that these patients are more likely to be subjected to radiotherapy. However, none of the studies is presently able to answer the question whether or not older age is per se a good prognostic factor. The present multi-institutional series is one of the largest studies on HDCT in children with centrally reviewed AT/RT. Striking clinical characteristics were the strong male predominance (3.75:1) and more importantly the high proportion of patients with metastatic disease (47%). Despite this negative selection, 2-year PFS (29±11%) and OS (50±12%) in our study are comparable to those reported by others. Radiotherapy appears to have contributed to the positive outcome in our long-term Bone Marrow Transplantation (2014) & 2014 Macmillan Publishers Limited

6 survivors. Complete CNS staging and central pathological review could be obtained in 84 and 95% of study patients. With 13 of 19 patients analyzed, this study also includes the most comprehensive data for SMARCB1 germline mutations in AT/RT patients treated with HDCT. Mutations were detected in 4/13 analyzed patients, which is an expected frequency; 22 interestingly, two of these patients are long-term survivors. Treatment recommendations were less consistently followed (for example, carboplatin/thiotepa was used as conditioning in 10/19 patients) reflecting the lack of evidence concerning optimal treatment of AT/RT. However, with increasing acceptance and recruitment numbers a stricter adherence to protocol guidelines is expected. CONCLUSION Although current evidence suggests that approximately one-third of patients with AT/RT can achieve sustained remissions after HDCT, the question which patient might eventually benefit from this approach cannot be answered based on available data. The present series are limited by small patient numbers, heterogeneous study populations and a non-randomized and retrospective study design. HDCT might be useful in patients with localized disease, good response to induction chemotherapy and no or small residual tumor following conventional therapy. In addition, the impact of each treatment modality on long-term survival remains to be elucidated. A meta-analysis of all available patients data might help to better define prognostic factors. The definitive role of HDCT, however, has to be evaluated in a randomized study. CONFLICT OF INTEREST The authors declare no conflict of interest. ACKNOWLEDGEMENTS We thank Petra Neumayer (EU-RHAB trial center Augsburg, Germany) for excellent data management and administrative support. EU-RHAB is supported by the Deutsche Kinderkrebsstiftung (German Childhood Cancer Foundation). REFERENCES 1 Burger PC, Yu IT, Tihan T, Friedman HS, Strother DR, Kepner JL et al. Atypical teratoid/rhabdoid tumor of the central nervous system: a highly malignant tumor of infancy and childhood frequently mistaken for medulloblastoma: a Pediatric Oncology Group study. Am J Surg Pathol 1998; 22: Woehrer A, Slavc I, Waldhoer T, Heinzl H, Zielonke N, Czech T et al. Incidence of atypical teratoid/rhabdoid tumors in children: a population-based study by the Austrian Brain Tumor Registry, Cancer 2010; 116: Rorke LB, Packer RJ, Biegel JA. Central nervous system atypical teratoid/rhabdoid tumors of infancy and childhood: definition of an entity. J Neurosurg 1996;85: von Hoff K, Hinkes B, Dannenmann-Stern E, von Bueren AO, Warmuth-Metz M, Soerensen N et al. Frequency, risk-factors and survival of children with atypical teratoid rhabdoid tumors (AT/RT) of the CNS diagnosed between 1988 and 2004, and registered to the German HIT database. Pediatr Blood Cancer 2011;57: Hilden JM, Meerbaum S, Burger P, Finlay J, Janss A, Scheithauer BW et al. Central nervous system atypical teratoid/rhabdoid tumor: results of therapy in children enrolled in a registry. J Clin Oncol 2004; 22: Buscariollo DL, Park HS, Roberts KB, Yu JB. Survival outcomes in atypical teratoid rhabdoid tumor for patients undergoing radiotherapy in a Surveillance, Epidemiology, and End Results analysis. Cancer 2012; 118: Dufour C, Beaugrand A, Le Deley MC, Bourdeaut F, André N, Leblond P et al. Clinicopathologic prognostic factors in childhood atypical teratoid and rhabdoid tumor of the central nervous system: a multicenter study. Cancer 2012; 118: Lafay-Cousin L, Hawkins C, Carret AS, Johnston D, Zelcer S, Wilson B et al. Central nervous system atypical teratoid rhabdoid tumours: the Canadian Paediatric Brain Tumour Consortium experience. Eur J Cancer 2012; 48: Tekautz TM, Fuller CE, Blaney S, Fouladi M, Broniscer A, Merchant TE et al. Atypical teratoid/rhabdoid tumors: improved survival in children 3 years of age and older with radiation therapy and high-dose alkylator-based chemotherapy. J Clin Oncol 2005; 23: Kordes U, Gesk S, Frühwald MC, Graf N, Leuschner I, Hasselblatt M et al. Clinical and molecular features in patients with atypical teratoid rhabdoid tumor or malignant rhabdoid tumor. Genes Chromosomes Cancer 2010; 49: Modena P, Sardi I, Brenca M, Giunti L, Buccoliero AM, Pollo B et al. Case report: long-term survival of an infant syndromic patient affected by atypical teratoidrhabdoid tumor. BMC Cancer 2013; 13: Chi SN, Zimmerman MA, Yao X, Cohen KJ, Burger P, Biegel JA et al. Intensive multimodality treatment for children with newly diagnosed CNS atypical teratoid rhabdoid tumor. J Clin Oncol 2009; 20: Zimmerman MA, Goumnerova LC, Proctor M, Scott RM, Marcus K, Pomeroy SL et al. Continuous remission of newly diagnosed and relapsed central nervous system atypical teratoid/rhabdoid tumor. J Neurooncol 2005; 72: Gidwani P, Levy A, Goodrich J, Weidenheim K, Kolb EA. Successful outcome with tandem myeloablative chemotherapy and autologous peripheral blood stem cell transplants in a patient with atypical teratoid/rhabdoid tumor of the central nervous system. J Neurooncol 2008; 88: Fidani P, De Ioris MA, Serra A, De Sio L, Ilari I, Cozza R et al. A multimodal strategy based on surgery, radiotherapy, ICE regimen and high dose chemotherapy in atypical teratoid/rhabdoid tumours: a single institution experience. J Neurooncol 2009; 92: Nicolaides T, Tihan T, Horn B, Biegel J, Prados M, Banerjee A. High-dose chemotherapy and autologous stem cell rescue for atypical teratoid/rhabdoid tumor of the central nervous system. J Neurooncol 2010; 98: Finkelstein-Shechter T, Gassas A, Mabbott D, Huang A, Bartels U, Tabori U et al. Atypical teratoid or rhabdoid tumors: improved outcome with high-dose chemotherapy. J Pediatr Hematol Oncol 2010; 32: e182 e Park ES, Sung KW, Baek HJ, Park KD, Park HJ, Won SC et al. Tandem high-dose chemotherapy and autologous stem cell transplantation in young children with atypical teratoid/rhabdoid tumor of the central nervous system. J Korean Med Sci 2012; 27: Sung KW, Lim DH, Lee SH, Yoo KH, Koo HH, Kim JH et al. Tandem high-dose chemotherapy and auto-sct for malignant brain tumors in children under 3 years of age. Bone Marrow Transplant 2013; 48: Gardner SL, Asgharzadeh S, Green A, Horn B, McCowage G, Finlay J. Intensive induction chemotherapy followed by high dose chemotherapy with autologous hematopoietic progenitor cell rescue in young children newly diagnosed with central nervous system atypical teratoid rhabdoid tumors. Pediatr Blood Cancer 2008; 51: Finlay J, Dhall G, Haley K, Erdreich-Epstein A, Gonzalez I, Allen J et al. Irradiationavoiding strategy for treatment of young children with central nervous system atypical teratoid/rhabdoid tumors: the head start experience, Pediatr Blood Cancer 2009; 53: 728 (abstract O.058). 22 Bourdeaut F, Lequin D, Brugières L, Reynaud S, Dufour C, Doz F et al. Frequent hsnf5/ini1 germline mutations in patients with rhabdoid tumor. Clin Cancer Res 2011; 17: Supplementary Information accompanies the paper on Bone Marrow Transplantation website ( & 2014 Macmillan Publishers Limited Bone Marrow Transplantation (2014)