The therapy of infantile malignant brain tumors: current status?

Transcription

1 Journal of Neuro-Oncology (2005) 75: Ó Springer 2005 DOI /s x Childhood Brain Tumors The therapy of infantile malignant brain tumors: current status? Chantal Kalifa and Jacques Grill Pediatric Department, Institut Gustave Roussy, 39 rue Camille Desmoulins, 94805, Villejuif ce dex, France Key words: atypical teratoid rhabdoid tumors, brain tumors, chemotherapy, choroid plexus carcinomas, ependymomas, infants, malignant gliomas, medulloblastomas, sequelae Summary Malignant brain tumors are not uncommon in infants as their occurrence before the age of three represents 20 25% of all malignant brain tumors in childhood [1]. Genetic predisposition to infantile malignant brain tumors are known in Gorlin syndrom for example who present with desmoplastic medulloblastoma in about 5% of the affected patients. In addition, sequelae from tumor and its treatment are more severe at this age [2]. Thus, malignant brain tumors represent a true therapeutic challenge in neuro-oncology. Before the era of modern imaging and modern neurosurgery these malignant brain tumors were misdiagnosed or could not benefit of the surgical procedures as well as older children because of increased risks in this age group. Since the end of the 80s, noninvasive imaging procedures produce accurate diagnosis of brain tumors and improvement in neurosurgery, neuroanesthesia and perioperative intensive care permit safe tumor resections or at least biopsies. Consequently, the pediatric oncologists are more often confronted with very young children who need a complementary treatment. Before the development of specific approaches for this age group, these children received the same kind of treatment than the older children did, but their survival and quality of life were significantly worse. The reasons of these poor results were probably due in part to the fear of late effects induced by radiation therapy, leading to decrease the necessary doses of irradiation which increased treatment failures without avoiding treatment related complications [3]. At the end of the 80s, pilot studies were performed using postoperative chemotherapy in young medulloblastoma patients. Van Eys treated 12 selected children with medulloblastoma with MOPP regimen and without irradiation; 8 of them were reported to be long term survivors [4]. Subsequently, the pediatric oncology cooperative groups studies have designed therapeutic trials for very young children with malignant brain tumors. Different approaches have been explored: Prolonged postoperative chemotherapy and delayed irradiation as designed in the POG (Pediatric Oncology Goup). Postoperative chemotherapy without irradiation in the SFOP (Socie te Franc aise d Oncologie Pe diatrique) and in the GPO (German Pediatric Oncology) studies. The role of high-dose chemotherapy with autologous stem cells transplantation was explored in different ways: High-dose chemotherapy given in all patients as proposed in the Head Start protocol High-dose chemotherapy given in relapsing patients as salvage treatment in the French strategy In the earliest trials, the same therapy was applied to all histological types of malignant brain tumors and whatever the initial extension of the disease. This attitude was justified by the complexity of the classification of all brain tumors that has evolved over the past few decades leading to discrepancy between the diagnosis of different pathologists for a same tumor specimen. Furthermore, it has become increasingly obvious that the biology of brain tumors in very young children is different from that seen in older children. However, in the analysis of these trials an effort was made to give the results for each histological groups, according to the WHO classification and after a central review of the tumor specimens. All these collected data have brought to an increased knowledge of infantile malignant brain tumors in terms of diagnosis, prognostic factors and response to chemotherapy. Furthermore a large effort was made to study long term side effects as endocrinopathies, cognitive deficits, cosmetic alterations and finally quality of life in long term survivors. Prospective study of sequelae can bring information on the impact of the different factors as hydrocephalus, location of the tumor, surgical complications, chemotherapy toxicity and irradiation modalities. With these informations it is now possible to design therapeutic trials devoted to each histological types, adapted to pronostic factors and more accurate treatment to decrease long term sequelae. Medulloblastoma/PNET The introduction of chemotherapy in patients with medulloblastomas is currently changing the philosophy of their treatment. The first publication of the Baby POG 1 study is considered as a milestone paper for the

2 280 treatment of brain tumors in children <3 years [5]. In this trial the most common tumors were medulloblastomas. Sixty-two children were treated with postoperative chemotherapy and delayed craniospinal irradiation performed after the age of 3. The 5-year PFS is 31.8 ± 8.3% and the 5-year overall survival is 39.7 ± 6.9%. Depending on the age at diagnosis, the delay in radiation of 1 or 2 years had no impact on survival. The strongest pronostic factor was the quality of the surgical resection : the 5-year survival was 60% for the 20 children who had a gross total resection and 32% for the 33 children who had an incomplete resection. The 5-year survival of the 13 patients who had had a gross total resection and no metastases at diagnosis was 69%. Most of the failures occurred during the first 6 months of chemotherapy with no progressive disease occurring after 2 years of therapy [6]. A following study comparing the Baby POG 1 dose regimen with a moderately dose-intensified version of the same regimen has been conducted. Preliminary results suggest no advantage for dose intensification (Strother personal communication). Forty-six children less than 18 months have been involved in the CCG study of postoperative chemotherapy «eight drugs in 1 day» regimen and proposed delayed irradiation. The 3-year PFS was 22 ± 6% (30% for those with gross total resection and no metastasis) but most the patients who had prolonged survivals did not receive irradiation following completion of chemotherapy [7]. A different approach have been taken in a pilot study from Philadelphia: 10 children months of age with localized medulloblastoma received a reduced dose radiation therapy with 18 Gy to the craniospinal axis, a dose of Gy to the posterior fossa and a chemotherapy consisting of weekly vincristine during irradiation and 8 cycles delivered every 6 weeks of the combination of vincristine, cisplatin, CCNU. There were 7 survivors and the overall survival at 6 years is 70±20% [8]. Others groups have attempted to eliminate radiation in young medulloblastoma patients. The first report was those of 12 medulloblastoma patients <3 years treated from 1976 to 1988 with postoperative MOPP regimen (nitrogen mustard, vincristine, procarbazine, prednisone); 8 of them are long term survivors and 6 of these 8 patients did not receive radiation [9]. The Australian New Zealand Childhood Cancer Group treated 16 medulloblastoma patients with a combination of cylophosphamide, vincristine etoposide; with a median follow-up of 25 months, 7 remained alive, none of them having received irradiation [10]. In the first German study SKK87 for medulloblastoma patients <3 years postoperative chemotherapy comprised procarbazine, ifosphamide, etoposide, high dose methotrexate and delayed irradiation; the 5-year survival rate was 50% in 30 children (Kuehl personal communication). In the HIT SKK 92 (which included high risk as well as standard risk patients), radiation was eliminated in children without evidence of disease after chemotherapy which consisted in cyclophosphamide, carboplatin, vincristine, etoposide, intravenous high dose (5 g/m 2 ) and intraventricular methotrexate; 45 of 62 eligible patients were treated according to the protocol and the 5-year PFS and OS were 55.5 ± 7.7% and 63.3 ± 7.4% respectively. Fourteen of 18 patients without postoperative residual tumor and without metastasis remained in CCR (PFS 77.8 ± 9.8%). In 14 patients with postoperative residue but no metastases the 5-year PFS and OS are 50 ± 13.4% and 56.3 ± 13.5%. PFS and OS were lower in M2 M3 patients (30.7 ± 12.8% and 34.6 ± 14.4%). In this study a correlation was found between the cumulative dose of intraventricular methotrexate and the risk of leukoencephalopathies (Rutkowski personal communication). In the Head Start I protocol, medulloblastoma patients received postoperative chemotherapy with vincristine, cisplatin, etoposide, cyclophosphamide and a high dose combination of carboplatin, etoposide and thiotepa with stem cells rescue; no irradiation was performed if the patients were in remission at the end of treatment. The 2-year EFS of the 12 medulloblastoma patients <3 years who entered this study was 38% [11]. The Head Start II protocol was designed for patients presenting with leptomeningeal dissemination; the postoperative regimen was intensified with high-dose methotrexate. The complete response rate observed in 21 medulloblastoma patients was 81%. The 3-year EFS was 49% (95% CI, 27 72%) [12]. The French BB-SFOP protocol was designed to deliver prolonged postoperative chemotherapy in young children with a malignant brain tumor and to avoid radiation therapy. Chemotherapy was given in alternating cycles of carboplatin and procarbazine, etoposide and cisplatin and vincristine and cyclophosphamide, over 16 months. No irradiation was given in patients unless they relapsed. In case of local progressive or recurrent disease, the recommended salvage treatment was high-dose busulfan (600 mg/m 2 ) and thiotepa (900 mg/m 2 ) with autologous stem cell support, followed by irradiation limited to the site of disease at the time of progression [13]. In 1995 the protocol was modified for high-risk patients. All patients with a postoperative local residue received the high-dose busulfan-thiotepa combination with autologous stem cells transplant followed by 50 Gy irradiation limited to the posterior fossa. Medulloblastoma patients with a leptomeningeal dissemination at diagnosis were enrolled into a repeated high-dose chemotherapy strategy (two courses of melphalan 100 mg/m 2 and one course of busulfan ( mg/m 2 ) and thiotepa ( mg/m 2 ); autologous peripheral blood stem cells transplant were performed after these three courses). Radiation therapy was administered at the dose of 50 Gy on the primary tumor bed (and not on the cranial axis). Eighty medulloblastoma patients have entered the BB-SFOP study. Fifty-eight patients have relapsed and 39/58 have received the salvage treatment using highdose busulfan-thiotepa and irradiation to the site of disease (5 of the metastatic patients, 12 of the patients with a post operative residue, 22 of the standard-risk patients). Four patients have received the sequential high-dose chemotherapy regimen at the time of a metastatic relapse.

3 281 Thirteen of 15 patients with metastatic medulloblastoma experienced a progressive disease during BBSFOP chemotherapy and the 5-year PFS for this group of patients is only 16%. Fifteen of the 17 patients with a postoperative residue progressed during the BBSFOP chemotherapy and the 5-year OS is 35%. For the group of 47 standard-risk patients 5-year PFS and OS are 32 ± 10% and 76 ± 10%, respectively. In this third group, the probability of prolonged complete remission was associated with complete surgical removal as mentioned in the operative report and not only the absence of residual tumor on the postoperative imaging (Grill personal communication). After 1995 six children with a postoperative local residue received the modified protocol; 5 of them are in first complete remission. Seventeen metastatic medulloblastoma patients were eligible for the repeated highdose chemotherapy strategy; the response rate to the two first courses of melphalan was 60%; 14/17 received the busulfan-thiotepa combination. Severe pulmonary toxicity leading to death was observed in the two first patients who received 600 mg/m 2 of busulfan and 900 mg/m 2 of thiotepa and led to decrease the doses in the following patients. The response rate to the busulfan-thiotepa course was 75%. Nowadays 5 patients (30%) are alive without disease with a median follow-up of 48 months (Valteau personal communication). Unexpected delayed neurotoxic effects of the busulfan-thiotepa combination and irradiation were observed: in 20% of the patients; transient gadolinium enhancements limited to the irradiated areas of the brain were observed several months after the end of treatment; in some patients this phenomenon was associated with a worsening of the neurological deficits. The recovery was observed in the majority of the cases within several months. The conclusion of the BBSFOP trial was that localized medulloblastoma patients can be cured with surgery and BBSFOP chemotherapy provided complete surgery can be carried out. On the contrary, this chemotherapy is ineffective for patients with incomplete surgery or metastasis at diagnosis. A major concern about these attempts to delay or eliminate radiation therapy was the efficacy of irradiation delivered as salvage treatment in case of disease progression. In the St. Jude experience salvage with radiation was possible in 5 out of 11 children who developed progressive disease during chemotherapy [14]. The French strategy using high dose busulfan and thiotepa and irradiation limited to the tumor bed for salvage treatment is an alternative to craniospinal irradiation which is highly effective for patients with a local residue or a local relapse [15]. Overall, 40 patients have received high dose busulfan-thiotepa and limited irradiation for a local residue or progression of a medulloblastoma. Overall survival at 5 years is 69% (OS is 100% for patients with local residue when treated initially with high dose chemotherapy, 67% for patients relapsing locally under chemotherapy and only 46% when treated at progression) (Ridola personal communication). These studies show that medulloblastoma in infants have the same pronostic factors than medulloblastomas in older children and can be divided as well in standard risk and high risk groups depending on the quality of the surgical resection and the presence of metastases at diagnosis. However the French study demonstrates that the criteria to define the extent of resection can be improved by considering the operative report rather than the postoperative imaging. Progression free survival of infants in these studies delaying or avoiding irradiation are inferior to those observed in older children, especially in case of incomplete resection or metastatic tumor. Nevertheless, 30 50% of the patients with a completely resected local tumor can be cured without irradiation. The overall survival rates in localized disease could however reach those observed in older children despite the omission or reduction of radiotherapy. Supratentorial PNET The group of nonmedulloblastoma PNET is a large basket of highly malignant tumors mainly located in the supratentorial area; they often develop in young children and leptomeningeal dissemination is frequent. Twenty-eight PNET patients were treated in the baby POG 1; among them 11 had a pinealoblastoma and were analyzed separately. The OS of the 17 children with a nonpineal PNET was 27% at 5 years. All the 11 pinealoblastomas failed chemotherapy; all children died in spite of irradiation delivered at the time of failure in 6 of them [16]. Same dismal results were observed with the 8 in 1 chemotherapy in the CCG trial: disease progression was observed in the 8 pinealoblastoma children [17]. However, in 11 children <18 months with supratentorial nonpinealoblastoma PNET treated with 8 in 1 and delayed or reduced irradiation the PFS was 55% [18] When craniospinal irradiation was delivered in addition to 8 in 1 chemotherapy in 12 supratentorial PNET (including pinealoblastomas) children aged 1 3 years, the 3-year PFS was 25 ± 13% ; the benefit of chemotherapy was not demonstrated in a randomized study including older children [19]. Twenty-five children with a supratentorial PNET were treated according to the BB-SFOP protocol [20]. After central histological review they were classified as 17 primitive neuroectodermal tumors NOS, 4 pinealoblastomas, 4 medulloepitheliomas. Four children presented with disseminated leptomeningeal disease; total resection was performed in 9 patients, subtotal in 9, partial in 3 and a biopsy in 4. Twenty-four patients have progressed or relapsed with a median time of 5.5 months. The 2- and 5-year survivals were 30 and 14% respectively. At the time of progression, only four patients were treated with high-dose busulfan-thiotepa and irradiation localized to the tumor bed; two of these four children are in continuous complete remission. These results show that these chemotherapy regimens are less effective in PNET than in medulloblastoma patients; local irradiation at least seems unavoidable. Further investigative studies must be conducted in this group of tumors.

4 282 Ependymomas Ependymomas represent the second common brain tumor in young children and are more often located in the posterior fossa than in the supratentorial area. Metastatic disease is rare. Phase II studies of single agents or combinations of drugs (including high dose regimens) show poor responses of ependymomas to chemotherapy [21]. However chemotherapy has often been used in an adjuvant setting after surgery and irradiation for ependymomas. Forty eight children <3 years with intracranial ependymomas entered the Baby POG 1 study. The 31patients who were <24 months at diagnosis received 2 years of chemotherapy followed by irradiation whereas the 17 patients aged months received chemotherapy for 1 year before irradiation. Unlike children with medulloblastomas in whom failures occurred early, children with ependymomas develop progressive disease over several years. Five year PFS and OS are 25% (±8.2%) and 40.5% (±7.2%) for the whole population. Five year survivals are 66% for children who had a gross total resection and only 25% for those with a subtotal resection. Furthermore, there was a significant difference in PFS according to the age at diagnosis as 5 year PFS were 12.7% (±8.4%) for children <24 months who had received 2 years of chemotherapy and 54.8% (±15%) in the months patients who received only 1 year of chemotherapy before radiotherapy. It was concluded that delaying irradiation adversely affected survival [22]. The 3-year PFS of the 15 infants with malignant ependymomas who received the 8 in 1 regimen according to the CCG protocol was 22% (Geyer 94). Ten patients were treated with the Head Start I Study and the 2 year PFS and OS are 30 and 60%, respectively [11]. Seventy-three children have been included in the initial report of the BBSFOP protocol to receive the 16-month postoperative chemotherapy without irradiation. In the event of relapse or progression, salvage treatment consisted in a second surgical procedure followed by local irradiation with or without second line chemotherapy. The 4-year PFS and OS were 22% (13 43%) and 59% (47 71%). Overall, 40% of the patients were alive having never received radiotherapy 2 years after the initiation of chemotherapy and 23% were still alive at 4 years without recourse to this modality. In the multivariate analysis, the two factors associated with a favorable outcome were a supratentorial tumor location and complete surgery. Overall survival at 4 years was 74% (59 86%) for patients in whom resection was radiologically complete and 35% (18 56%) for the patients with incomplete resection. High-dose busulfanthiotepa with autologous stem cells rescue was administered in 16 progressing patients: no tumor regression was observed with this procedure [23]. The conclusion of this study was that a significant proportion of children with ependymoma can avoid radiotherapy with prolonged adjuvant chemotherapy. From this French trial it was concluded that irradiation can be deferred safely in children with radiologically proven gross total resection to the time of progression. Because of the poor results of chemotherapy in ependymomas, new modalities of irradiation are currently explored. A phase II trial of conformal radiation therapy for localized childhood ependymoma was done to determine whether the irradiated volume could be reduced to decrease CNS-related side effects without diminishing the rate of disease control. Eighty-eight children (median age 2.85 years, 15 <18months) have received Gy to the gross tumor volume and a margin of 10 mm. The 3-year PFS was 74.7% (±5.7%). The cumulative incidence of local failure at 3 years was 14.8%. The conclusion is that limited-volume irradiation achieves high rates of disease control in pediatric ependymoma [24]. At least, one must mention that few single institutions reports have shown that some ependymoma patients can be cured by surgery only [25]. This is more likely to occur with low grade supratentorial tumors and in children >3 years of age. However, as two thirds of the patients will relapse using current chemotherapy protocols, a more efficient adjuvant treatment is mandatory for the whole population. Malignant gliomas High grade gliomas are known to have a poor outcome; they represent the third most common malignant brain tumor type in children less than 3 years of age and they develop mainly during the first year of life. Eighteen high grade glioma children entered the Baby POG 1 protocol (6 glioblastomas multiforme, 9 unclassified malignant gliomas and 3 anaplastic astrocytomas). Thirteen were younger than 12 months at diagnosis. PFS and OS at 5 years were 43% (±23%) and 50% (±14%). No failures occurred after 2 years and 4 patients among the survivors have received no irradiation [26]. Same surprising data have been observed with the French BBSFOP protocol; 21 patients less than 5 years have received this postoperative chemotherapy for a high grade glioma (11 anaplastic astrocyromas, 7 anaplastic oligodendrogliomas, 3 glioblastomas multiforme). At 5 years the PFS and OS were 31% (16 52%) and 54% (34 74%), respectively. The age appears to be a pronostic factor as 6/11 children less than 2 years versus only 1/10 more than 2 years are long term survivors and in complete remission. Six long term survivors did not receive irradiation (Dufour personal communication). Both studies suggest a particular behavior for infants with high grade gliomas. Atypical teratoid rhabdoid tumors Atypical teratoid rhabdoid tumors (ATRT) are extremely rare and aggressive tumors developing in early childhood. Central nervous system is the most frequent location of these tumors that have been originally described in the kidneys. Cytogenetic abnormalities of 22q11.2 are present in about 75% of these tumors which present complex histologic features [27,28]. Molecular

5 283 genetic has identified a rhabdoid suppresor gene (INI1/ hsnf5) at 22q11.2. Before they were recognized as a special entity, these tumors were treated as medulloblastomas and time course from diagnosis to death was about 1 year (Burger am j pathol 98). More aggressive therapy have been used leading to prolonged survival in some patients. Fortytwo ATRT children have been enrolled in a US registry; they have received different treatments but 14 patients show no evidence of disease with a median follow-up of 46 months. The analysis of these data give the following indications [30]: The quality of the surgical resection is a major pronostic factor and maximal surgery is recommended ATRT are chemosensitive tumors when platin derivatives and alkylating agents are used (response rate 50%) Radiotherapy limited to the tumor bed is recommended although some patients are long term survivors without having received any radiation Intrathecal chemotherapy is recommended, especially in metastatic patients High dose chemotherapy with autologous stem cells transplantation has to be investigated in prospective studies. Pilot studies are currently conducted in pediatric oncology groups. Experience gathered in older children treated at St. Jude with craniospinal irradiation plus high dose alkalyting agents suggest that radiotherapy may be an important component of the treatment (Tekautz personal communication). Choroid plexus carcinomas Choroid plexus carcinomas represent about one third of all choroid plexus tumors which comprise 1% of all pediatric CNS tumors. An analysis of the literature demonstrate the importance of a gross total resection of these tumors. This supports a second surgery if the first one was incomplete. Adjuvant therapy was used in almost all the cases, mainly radiation in the past and in older children and more recently postoperative chemotherapy in younger patients. The role of radiation to increase the probability of survival in patients with a gross total resection is controversial [31,32]. The number of patients who received an adjuvant chemotherapy is too small to conclude definitively on the value of chemotherapy to prolong survival but long term survivors were reported without irradiation [33,34]. Nevertheless there is some evidence of the efficacy of multiagent chemotherapy on mesurable disease [35]. The interest of postoperative chemotherapy on residual tumor in order to facilitate a second surgery by decreasing the vascularity of the tumor has been emphasized [3]. Mutations of the hsnf/ini1 tumor suppressor gene have been found in choroid plexus carcinomas as well as in ATRT. Further studies will hopefully clarify the relationship between both tumors and help to define therapeutic indications [36]. In conclusion Besides improving survivals the aim of all these trials performed in young children was to decrease the adverse effects of radiation therapy, especially on intellectual development. This means that long term evaluations must be performed. There are some preliminary data which show the positive impact of these new strategies on the cognitive issues [37]. Nevertheless one must be aware not to replace radiation toxicity by chemotherapy toxicity. The role of high dose methotrexate in the incidence of leucoencephalopathies in the German study is clear. In the French strategy the neurotoxicity of the combination of high dose busulfan and thiotepa followed by radiation has also been observed in some patients. The benefit of reducing the cranispinal dose of radiation to 18 Gy in young children has been shown in one study [8] but not confirmed by a second one [17]. The incidence of second tumors developing after a first brain tumor in young patients has been reported [39] but this risk does not appear superior to the one observed after irradiation [40]. Future development in the therapy of these diseases will certainly benefit of better knowledge of the histology and biology of these brain tumors. Young children have taught us new avenues for the treatment of brain tumors that could benefit to older ones also. References 1. Rickert CH, Paulus W: Epidemiology of central nervous system tumors in childhood and adolescence based on the new WHO classification. Childs Nerv Syst 17: , Suc E, Kalifa C, Brauner R, Habrand JL, Terrier-Lacombe MJ, Vassal G, Lemerle J: Brain tumors under the age of three. The price of survival. A retrospective study of 20 long term survivors. Acta Neurochir (Wien) 106: 93 98, Duffner PK, Cohen ME, Myers MM, Heise HW: Survival of children with brain tumors: SEER Program Neurology 36: , Van Eys J, Cangir A, Coody D, Smith B: MOPP regimen as primary chemotherapy for brain tumors in infants. J Neurooncol 3: , Duffner PK, Horowitz ME, Krischer JP, Friedman HS, Burger PC, Cohen ME, Sanford RA, Muhler RK, James HE, Freeman CR, Seidel FE, Kun LE: Postoperative chemotherapy and delayed radiation in children less than three years of age with malignant brain tumors. N Engl J Med 328: , Duffner PK, Horowitz ME, Krischer JP, Burger PC, Cohen ME, Sanford RA, Friedman HS, Kun LE: The treatment of malignant brain tumors in infants and very young children: an update of the Pediatric Oncology Group experience. Neuro-oncol 1: , Geyer JR, Zeltzer PM, Boyett JM, Rorke L, Stanley P, Albright AL, Wisoff JH, Milstein JM, Allen JC, Finlay JL, Ayers GD, Shurin SB, Stevens K, Bleyer WA: 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 12: , 1994

6 Goldwein JW, Radcliffe J, Johnson J, Moshang T, Packer RJ, Sutton LN, Rorke LB, D Angio GJ: Updated results of a pilot study of low dose craniospinal irradiation plus chemotherapy for children under five with cerebellar primitive neuroectodermal tumors (Medulloblastoma). Int J Radiat Oncol Biol Phys 44: , Ater JL, Eys Jvan, Woo SY, Moore B, Copeland DR, Bruner J: MOPP chemotherapy without irradiation as primary postsurgical therapy for brain tumors in infants and young children. J Neurooncol 32: , White L, Kellie S, Gray E, Toogood I, Waters K, Lockwood L, Macfarlane S, Johnston H: 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 20: , Mason WP, Grovas A, Halpern S, Dunkel IJ, Garvin J, Heller G, Rosenblum M, Gardner S, Lyden D, Sands S, Pucetti D, Lindsey K, Merchant TE, O Malley B, Bayer L, Petriccione MM, Allen J, Finlay JL: Intensive chemotherapy and bone marrow rescue for young children with newly diagnosed malignant brain tumors. J Clin Oncol 16: , Chi SN, Gardner S, Levy AS, Knopp EA, Miller DC, Wisoff JH, Weiner HL, Finlay JL: Feasibility and response to induction chemotherapy intensified with high dose methotrexate for young children with newly diagnosed high risk disseminated medulloblastoma. J Clin Oncol 22: , Kalifa C, Hartmann O, Demeocq F, Vassal G, Couanet D, Terrier-Lacombe MJ, Valteau D, Brugieres L, Lemerle J: High-dose busulfan and thiotepa with autologous bone marrow transplantation in childhood malignant brain tumors: a phase II study. Bone Marrow Transpl 9: , Gajjar A, Mulhern RK, Heideman RL, Sanford RA, Douglass EC, Kovnar EH, Langston JA, Jenkins JJ, Kun LE: Medulloblastoma in very young children: outcome of definitive craniospinal irradiation following incomplete response to chemotherapy. J Clin Oncol 12: , Dupuis-Girod S, Benhamou E, Doz F, Mechinaud F, Bouffet E, Coze C, Kalifa C: Will high dose chemotherapy followed by autologous bone marrow transplantation supplant cranio-spinal irradiation in young children treated for medulloblastoma?. J Neurooncol 27: 87 98, Duffner PK, Cohen ME, Sanford RA, Horowitz ME, Krisher JP, Burger PC, Friedman HS, Kun LE: Lack of efficacy of postoperative chemotherapy and delayed radiation in very young children with pinealoblastoma. Med Pediatr Oncol 25: 38 44, Jackaki RI, Zeltzer PM, Boyett AL, Albright AL, Allen JC, Geyer JR, Rorke LB, Stanley P, Stevens KR, Wisoff J, McGuire-Cullen P, Milstein J, Packer RJ, Finlay JL: Survival and prognostic factors following radiation and/or chemotherapy for primitive neuroectodermal tumors of the pineal regions in infants and children: a report of the Children s Cancer Group. J Clin Oncol 13: , Geyer JR, Zeltzer PM, Boyett JM, Rorke LB, Stanley P, Albright AL, Wisoff JH, Milstein JM, Allen JC, Finlay JL, Ayers GD, Shurin SB, Stevens K, Bleyer WA: Survival of infants with primitive neuroectodermal tumors or malignant ependymomas of the CNS treated with eight drugs in 1 day: A report from the Children s Cancer Group. J Clin Oncol 12: , Cohen BH, Zeltzer PM, Boyett JM, Geyer JR, Allen JC, Finlay JL, McGuire-Cullen P, Milstein JM, Rorke LB, Stanley P, Stehbens JA, Shurin SB, Wisoff J, Stevens K, Albright AC: prognostic factors and treatment results for supratentorial primitive neuroectodermal tumors in children using radiation and chemotherapy: a Children s Cancer Group randomized trial. J Clin Oncol 13: , Marec-Berard P, Jouvet A, Thiesse P, Kalifa C, Doz F, Frappaz D: Supratentorial embryonal tumors in children under 5 years of age: an SFOP study of treatment with postoperative chemotherapy alone. Med Pediatr Oncol 38: 83 90, Grill J, Chastagner P, Kalifa C: Childhood ependymoma: a systematic review of treatment options and strategies. Paediatr Drugs 5: , Duffner PK, Krischer JP, Sanford RA: Prognostic factors in infants and very young children with intracranial ependymomas. Pediatr Neurosurg 28: , Grill J, Le Deley MC, Gambarelli D, Raquin MA, Couanet D, Pierre-Kahn A, Habrand JL, Doz F, Frappaz D, Gentet JC, Edan C, Chastagner P, Kalifa C: 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 19: , Merchant TE, Mulhern RK, Krasin MJ, Kun LE, Williams T, Li C, Xiong X, Khan RB, Lustig RH, Boop FA, Sanford RA: 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 22: , Hukin J, Epstein F, Lefton D, Allen J: treatment of intracranial ependymoma by surgery alone. Pediatr Neurosurg 29: 40 45, Duffner PK, Krischer JP, Burger PC, Chen ME, Backstrom JW, Horowitz ME, Sanford RA, Friedman HS, Kun LE: Treatment of infants with malignant gliomas: the Pediatric Oncology Group experience. J Neurooncol 28: , Rorke LB, Packer RJ, Biegel JA: Central nervous system atypical teratoid/rhabdoid tumors of infancy and childhood: definition of an entity. J Neurosurg 85: 56 65, Biegel JA, Zhou JY, Rorke LB, Stenstrom C, Wainwright LM, Fogelgren B: Germ-line and acquired mutations of INI1 in atypical teratoid and rhabdoid tumors. Cancer Res 59: 74 79, Burger PC, Yu IT, Tihan T, Friedman HS, Strother DR, Kepner JL, Duffner PK, Perlman EJ: 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 22: , Hilden JM, Meerbaum S, Burger P, Finlay J, Janss A, Scheithauer BW, Walter AW, Rorke LB, Biegel JA: Central nervous system atypical teratoid/rhabdoid tumor: results of therapy in children enrolled in a registry. J Clin Oncol 22: , Wolff JEA, Sajedi M, Coppes MJ, Anderson RA, Ageler RM: Radiation therapy and survival in choroid plexus carcinoma. Lancet 353: 2126, Fitzpatrick LK, Aronson LJ, Cohen KJ: Is there a requirement for adjuvant therapy for choroid plexus carcinoma that has been completely resected?. J Neurooncol 57: , Duffner PK, Kun LE, Burger PC, Horowitz ME, Cohen ME, Sanford RA, Krischer JP, Mulhern RK, James HE, Rekate HL: Postoperative chemotherapy and delayed radiation in infants and very young children with choroid plexus carcinomas. The Pediatric Oncology Group experience. Pediatr Neurosurg 22: , Berger C, Thiesse P, Lellouch-Tubiana A, Kalifa C, Pierre-Kahn A, Bouffet E: Choroid plexus carcinomas in childhood: clinical features and prognostic factors. Neurosurgery 42: , Greenberg ML: Chemotherapy of choroid plexus carcinoma. Childs Nerv Syst 15: 571 7, Razzaq AA, Cohen AR: Neoadjuvant chemotherapy for hypervascular malignant brain tumors in childhood. Pediatr Neurosurg 27: , Sevenet N, Lellouch-Tubiana A, Schofield D, Hoang-Xuan K, Gessler M, Birnbaum D, Jeanpierre C, Jouvet A, Delattre O: Spectrum of hsnf5/ini1 somatic mutations in human cancer and genotype-phenotype correlations. Hum Mol Genet 8 : , Sands SA, van Gorp WG, Finlay JL: Pilot neuropsychological findings from a treatment regimen consisting of intensive chemotherapy and bone marrow rescue for young children with newly diagnosed malignant brain tumors. Childs Nerv Syst 14: , Duffner PK, Krischer JP, Horowitz ME, Cohen ME, Burger PC, Friedman HS, Kun LE: Second malignancies in young children with primary brain tumors following treatment with prolonged postoperative chemotherapy and delayed irradiation: a Pediatric Oncology Group study. Ann Neurol 44: , 1998

7 Broniscer A, Ke W, Fuller CE, Gajjar A, Kun LE: Second neoplasms in pediatric patients with primary central nervous system tumors: the St Jude Children s Research Hospital experience. Cancer 100: , 2004 Address for offprints: Chantal Kalifa, Pediatric Department, Institut Gustave Roussy, 39 rue Camille Desmoulins Villejuif cédex France; Tel.: ; Fax: ; kalifa@igr.fr