Atypical Teratoid/Rhabdoid Tumors (ATRT): Improved Survival in Children 3 Years of Age and Older With Radiation Therapy and High-Dose Alkyla
http://www.100md.com
《临床肿瘤学》
the Department of Hematology-Oncology, Department of Pathology, Department of Radiological Sciences, Department of Biostatistics, and Department of Developmental Neurobiology, St Jude Children's Research Hospital, Memphis, TN
Texas Children's Hospital, Houston TX
ABSTRACT
PATIENTS AND METHODS: Primary tumor samples from patients diagnosed with ATRT at SJCRH between July 1984 and June 2003 were identified. Pathology review included histologic, immunohistochemical analysis, and fluorescence in situ hybridization for SMARCB1 (also known as hSNF5/INI1) deletion. Clinical records of patients with pathologic confirmation of ATRT were reviewed.
RESULTS: Thirty-seven patients were diagnosed with ATRT at SJCRH during the 19-year study interval. Six patients were excluded from this clinical review based on pathologic or clinical criteria. Of the remaining 31 patients, 22 were younger than 3 years. Posterior fossa primary lesions and metastatic disease at diagnosis were more common in younger patients with ATRT. All patients underwent surgical resection; 30 received subsequent chemotherapy. The majority of patients aged 3 years or older received postoperative craniospinal radiation. Two-year event-free (EFS) and overall survival (OS) of children aged 3 years or older (EFS, 78% + 14%; OS, 89% ± 11%) were significantly better than those for younger patients (EFS, 11% ± 6%; OS, 17% ± 8%); EFS, P = .009 and OS, P = .0001. No other clinical characteristics were predictive of survival. Three of four patients 3 years or older with progressive disease were successfully rescued with ifosfamide, carboplatin, and etoposide therapy.
CONCLUSION: Children presenting with ATRT before the age of 3 years have a dismal prognosis. ATRT presenting in older patients can be cured using a combination of radiation and high-dose alkylating therapy. Older patients with relapsed ATRT can have salvage treatment using ICE chemotherapy.
INTRODUCTION
The diagnosis of ATRT relies predominantly on morphologic and immunohistochemical criteria (IHC).8 Chromosome 22q11 abnormalities are frequently seen in ATRT and have proved valuable for diagnostic purposes. Chromosome 22q11.2 harbors the putative tumor suppressor gene SMARCB1 (also known as hSNF5/INI1), a component of an SWI/SNF ATP-dependent chromatin-remodeling complex that likely controls the expression of certain target genes involved in cell cycle regulation.9-11 Seventy-five percent of ATRTs contain homozygous deletions of SMARCB1, or loss of one allele and mutation of the other copy of the gene.12-15 Published series of ATRT report an overall 2-year survival rate of less than 15% for children aged younger than 3 years at diagnosis. Therapeutic approaches to treatment of ATRT have included a variety of postoperative chemotherapy regimens with or without radiation therapy (RT).2,3,16 Although RT appears most effective when administered early in the treatment of ATRT,2 the unacceptable sequelae associated with cranial irradiation in infants and young children precludes the use of this modality in most patients with this disease.2 For the first time we present the clinical characteristics and survival correlates from our institutional series of patients with newly diagnosed ATRT and compare the infants and young children (< 3 years) with older children (≥ 3 years).
PATIENTS AND METHODS
Clinical Review
Once patients with the diagnosis of ATRT were identified, a detailed review of the clinical data was performed to ascertain the presenting features, degree of surgical resection, chemotherapy regimen, RT dose, and volume.
Pathology, Immunohistochemistry, and Fluorescence In Situ Hybridization
All available tumor material from patients receiving an institutional diagnosis of ATRT between July 1984 and June 2003 was reviewed by a single neuropathologist (C.F.). Histopathologic review included morphologic assessment of ATRT characteristics according to WHO criteria and immunohistochemical (IHC) analysis. IHC was performed using a panel of six antibodies including epithelial membrane antigen (EMA), smooth muscle actin (SMA), vimentin, glial fibrillary acidic protein (GFAP), keratin, and synaptophysin. The specific immunohistochemical stain(s) employed for a given tumor varied and was dependent on the histologic appearance of the tumor and amount of tissue available.
Tissue microarrays (TMA) were constructed from formalin-fixed, paraffin-embedded archival tissue blocks using a tissue arrayer (Beecher Instruments, Silver Spring, MD). Two to 4 tissue cores (1.0-mm diameter) from histologically representative areas of tumor were included from each case. Five micrometer-thick sections were then cut from the TMA and mounted on poly-L-lysine-coated slides for dual-color fluorescence in situ hybridization (FISH) experiments as previously reported.9 Bacterial artificial chromosome (BAC)–derived probes targeting SMARCB1 (22q11.23; contiguous gene sequence of CTD-2034E7 and CTD-2355C2, Invitrogen, Huntsville, AL) and PANX2 (control for chromosome 22 ploidy-22q13.3; RPCI3-402G11, Children's Hospital Oakland Research Institute, Oakland, CA) were labeled in rhodamine and fluorescein, respectively, and diluted 1:50 in DenHyb hybridization buffer (Insitus Laboratories, Albuquerque, NM) for dual-target hybridization. The probe and target were simultaneously denatured at 90°C for 13 minutes, followed by overnight hybridization at 37°C in a humidified chamber. DAPI (1.0 μg/mL; Insitus Laboratories) was used as a nuclear counterstain, and the sections were viewed under a Nikon E800 fluorescent microscope with appropriate filters (Nikon Instruments, Melville, NY).
Sections showing sufficient hybridization efficiency were considered informative, and 100 to 200 intact nonoverlapping nuclei were scored for the number of fluorescent signals by two reviewers (C.F., J.D.). Cutoffs for abnormalities and/or deletions were based on counts from non-neoplastic control specimens (normal brain from autopsy cases) for each probe. Interpretation of deletion required more than 37% of tumor nuclei containing 1 (hemizygous) or zero (homozygous) red SMARCB1 signals (mean + 3 SDs in controls); all cases showing homozygous deletion retained demonstrable SMARCB1 signals within normal tissue components (endothelial cells). As nuclei with more than two signals were rarely seen in non-neoplastic controls, polysomies (gains) were arbitrarily defined as more than 5% nuclei containing three or more signals.
Images were captured using a high-resolution black and white COHU CCD camera and the CytoVision basic workstation (Applied Imaging, Santa Clara, CA). A Z-stack motor allowed for sequential DAPI (one level), FITC (16 levels), and rhodamine (16 levels) filter settings to be captured, and the resulting images were reconstructed with blue, green, and red pseudocolors.
Statistical Analysis
Overall survival (OS) was measured from the date of initial diagnosis of ATRT to the date of death or last contact. Event-free survival (EFS) was measured from the date of diagnosis to the first date of progressive disease for those who progressed, to the date of death for those patients who died without progressive disease or to the date of last contact for children who survived in the absence of treatment failure. Distributions of EFS and OS were estimated using the method of Kaplan and Meier.17 Associated SEs were calculated as suggested by Peto et al.18 Differences in OS and EFS distributions in relation to age at diagnosis were assessed by exact log-rank tests.19 In comparison with survival estimates in relation to other risk factors, stratification was based on age at diagnosis; stratified exact log-rank tests were also used. Only two-sided P values are reported. Exact tests were performed using the StatXact software package (Cytel Software Corp, Cambridge, MA).
RESULTS
Pathology and Molecular Review
There was a broad range of histopathologic and immunohistochemical features present in the tumors from this patient cohort. In addition to a primitive neuroectodermal component composed of generally patternless sheets of round blue cells, tumor samples from all patients harbored variable numbers of "rhabdoid cells" with abundant cytoplasm often containing eosinophilic filamentous inclusions, eccentrically situated nuclei with open chromatin, and large nucleoli. IHC analysis of 31 ATRTs demonstrated strong cytoplasmic positivity with EMA, the majority additionally showing polyphenotypic immunoprofiles consistent with ATRT including positivity for SMA, vimentin, GFAP, keratin, and synaptophysin, in decreasing order of frequency (Table 1). 8,9,20
Twenty-one tumor samples from 20 patients were available for FISH analysis (Table 1). Sixteen cases (76%) contained deletion of SMARCB1. Homozygous SMARCB1 deletion was detected in five cases (24%); the average percentage of cells showing homozygous deletion in these five samples was 84% (range, 67% to 92%). All five tumors that displayed homozygous deletion also displayed evidence of concomitant monosomy 22, with only a single PANX2 signal per cell (average percentage of cells, 51%; range, 34% to 80%). These data suggest that in these cases homozygous deletion occurred as a result of monosomy 22 and associated interstitial deletion of the second SMARCB1 allele. Eight cases (38%) showed deletion of a single copy of SMARCB1/hemizygous deletion (average percentage of cells, 52%; range, 39% to 93%). A significant proportion of these cases showed a larger region of loss on chromosome 22; up to 63% (range, 50% to 71%) of INI1-deleted cells displaying single red and green signals (monosomy) and a lesser percentage had only hemizygous loss of INI1 (single red INI1 and two green PANX2 signals per cell). Monosomy 22 alone was the predominant alteration in three cases (14%) with the majority of tumor cells involved (average, 82%; range, 76% to 86%). Only five cases (24%) had no demonstrable alterations of chromosome 22 detected by the current FISH assay. In one patient (patient 5), FISH analysis on tumor obtained at diagnosis did not detect any INI1 abnormality. The child subsequently developed recurrent disease; evaluation of the recurrent tumor demonstrated an INI1 hemizygous deletion by FISH.
ATRT Patient Cohort: Demographic/ Clinical Features
Twenty-two patients (71%) were younger than 3 years at diagnosis with a median age of 1 year (range, 0.2 to 2.7 years). Sixteen patients younger than 3 years old at diagnoses were male; 16 were white, four were African-American, and two were Hispanic. The median age at diagnosis of patients 3 years and older was 3.9 years (range, 3.3 to 7.4 years) and included three males; six were white, two were African Americans, and one was Hispanic.
Disease extent was evaluated by computed tomography and/or magnetic resonance (MR) imaging of the brain and spine after tumor resection and, Tc99 bone scintigraphy. CSF cytology was examined in all cases.
RT was delivered either to the entire neuraxis or locally. The dose of craniospinal irradiation (CSI) administered was determined by the patient's age, postoperative residual disease, and the presence of metastatic disease. Patients 3 years or older with less than 1.5 cm2 postoperative residual disease and no evidence of neuraxis dissemination received 2,340 cGy to the neuraxis whereas all other patients 3 years or older were treated to a neuraxis dose of 3,600 cGy. The primary site of disease or tumor bed was treated to a dose of 5,580 cGy using image-guided techniques.21 Patients who were treated with local-only RT had image-guided delivery of their treatment to cumulative doses selected by the treating physicians.
Primary tumors originated from the posterior fossa (n = 14, 45%), the supratentorial region (n = 16; 52%), and spine (n = 1). Posterior fossa was the most common primary tumor site among patients younger than 3 years (n = 12; 55%). In contrast, supratentorial primary tumors were predominant in the older patients (n = 7; 78%); only two (22%) tumors originated in the posterior fossa. Metastatic disease was present in six patients (27%) at diagnosis (M1 = 2, M2 = 2, M3 = 2); one patient (patient 17) was found to have an asymptomatic renal rhabdoid tumor. All patients with initial metastatic disease were younger than 3 years at diagnosis (Table 2).
Treatment for all patients included surgical resection of the tumor (Table 2). Of the 31 patients, 21 (68%) with ATRT underwent gross total or near total tumor resection (GTR/NTR); 14 of these patients were younger than 3 years. Ten patients (32%) had subtotal resections (STR), eight patients were younger than 3 years at diagnosis. Treatment administered after surgical excision varied depending on patient age (Table 3). All patients younger than 3 years were treated with chemotherapy, three also received RT or CSI. Seven of the nine patients 3 years or older were treated with RT and chemotherapy as outlined in Tables 3 and 4.
Therapeutic Response and Survival
No chemotherapy-related deaths occurred. Of the 22 patients younger than 3 years at diagnosis, 18 experienced recurrent or progressive disease at a median of 0.4 years from diagnosis (range, 0.1 to 0.7 years). Recurrence was local in 13 (72%), distant in 3 (17%), and both local and distant in 2 (11%). Thirteen of the 18 patients with relapsed disease were treated with chemotherapy (n = 4), RT (n = 4), or a combination of chemotherapy and RT (n = 5; Table 3). The five patients who did not receive further treatment experienced a median survival of 0.3 years (range, 0 to 0.3 years). Salvage treatment consisting of chemotherapy only was administered to four patients. Treatment included alkylator-based chemotherapy (cyclophosphamide/cisplatin/vincristine/etoposide; carboplatin/ifosfamide/etopside (ICE); high-dose cyclophosphamide and topotecan followed by autologous hematopoietic stem-cell reconstitution; and oxaliplatin/topotecan). The median survival was 0.3 years (range, 0.3 to 0.5 years).
Radiation only was employed in the treatment of four patients with recurrent disease, three of four patients had CSI (3,300 to 3,960 cGy), one received local RT (2,100 cGy). The median survival of these patients was 0.4 years (range, 0 to 0.8 years). Finally, five patients were treated with RT and chemotherapy after recurrence was detected. Radiation was administered before the chemotherapy in four of these children. Two were treated with CSI (2,500 and 3,960 cGy) and three received local RT (5,040 to 5,580 cGy). Chemotherapy in this group included two patients who received high-dose chemotherapy with autologous hematopoietic stem cell reconstitution (one of whom was treated with intrathecal methotrexate) in combination with an alkylator-based chemotherapy regimen. One patient received cisplatin and etoposide, and one patient was treated with a topoisomerase I inhibitor. The median survival in this group was 0.6 years (range, 0.3 to 1.4 years). All patients younger than 3 years with recurrent disease died within 1.5 years of progression with a median survival of 0.4 years (range, 0.06 to1.4 years).
Four of the 22 patients in the younger cohort did not develop recurrent or progressive disease. One patient (patient 16) died of respiratory failure (secondary to complications of surgery at the time of tumor resection) 2.5 months from diagnosis without evidence of progressive disease. The remaining three patients are alive without recurrent disease. One patient (patient 18) who received chemotherapy only electively terminated treatment at 6 months from diagnosis and is without evidence of disease 10 months from diagnosis. Two patients were treated with surgical resection, chemotherapy, and RT; they are alive 4.8 and 5.8 years from diagnosis (patients 7 and 6, respectively).
In contrast, of the nine patients 3 years of age or older at diagnosis, five remain free of disease at a median of 2.2 years (range, 0.8 to 4.1 years) from diagnosis. Four have experienced disease recurrence at a median of 1.2 years from diagnosis (range, 0.2 to 3.9 years; Table 3); recurrence was local in two and both local and distant in two. Two of these patients (patients 23 and 24) did not receive any RT as part of their initial therapy. Three of the four patients who experienced disease recurrence are alive (two with no evidence of disease; one alive with disease) at 0.6, 1.5, and 9.5 years after five to eight cycles of ifosphamide, carboplatin, etoposide, and RT in two of the three (local, 1; CSI, 1).
EFS and OS estimates for the entire cohort of 31 patients at 2 years were 31% ± 9% and 40% ± 10%, respectively. The median length of follow-up for the group younger than 3 years was 0.7 years (range, 0.2 to 5.8 years), and their 2-year EFS and OS estimates were 11% ± 6% and 17% ± 8%, respectively (Figs. 1A and B). For the group 3 years and older at diagnosis, the median length of follow-up was 2.8 years (range, 0.5 to 9.8 years); the 2-year EFS estimate was 78% ± 14%, and the 2-year OS estimate was 89% ± 11% (Figs. 1A and 1B). Given the striking association between age and EFS (P = .009; Fig 1A) and OS (P = .0001; Fig 1B), further analyses were stratified based on age at diagnosis. Primary tumor location was not predictive of EFS (P = .41). Similarly, EFS was not correlated with M-stage at diagnosis (P = .95), extent of the initial tumor resection (P = .096), or abnormalities in chromosome 22 (P = .82). Separate analysis of the two age groups, however, suggested an association between the extent of surgical resection (GTR/NTR v STR) and OS estimates (P = 0.053; data not shown).
Survival estimates for those patients initially treated with chemotherapy alone were compared with those for patients treated with chemotherapy and RT. In this analysis, we did not consider the chemotherapeutic agents used or doses administered, the extent and dose of RT, or the sequence in which these treatments were given. For the 21 patients treated with chemotherapy, the 1-year EFS estimate was 0%, the 1-year OS estimate was 42% ± 11%, and a 2-year OS estimate was 12% ± 7% (Figs. 2A and 2B). In contrast, the 2-year EFS and OS estimates for the 10 patients treated with chemotherapy and RT (three from the younger group and seven from the older group) were both 90% ± 10% (Figs. 2A and 2B). The correlation between treatment administered and EFS (P = .0003) and OS (P = .007) in our patient cohort with RT associated with improved survival. Therapy received was dependant on age, disease location, and disease extent, and thus the survival differences observed are likely attributable to multiple confounding factors. It is notable, however, only two long-term survivors (> 24 months from diagnosis) in children younger than 3 years at diagnosis from the current cohort both received RT as part of their up-front therapy, before the development of progressive disease.
DISCUSSION
Molecular and cytogenetic analysis initially implicated abnormalities of 22q11.2 in the pathogenesis of both ATRT and malignant rhabdoid tumor (MRT)12,13 and subsequently, the SMARCB1 gene was identified in this region.10 As part of the SWI/SNF chromatin remodeling complex, SMARCB1 functions as a DNA-binding protein, facilitating SWI/SNF-mediated cotransactivation of genes involved in cell cycle regulation.13,25,26 Homozygous deletions or truncation mutations of the SMARCB1gene abrogate SWI/SNF-imposed cell-cycle arrest11,13 and have been identified in up to 75% of ATRTs evaluated.9,13,27,28 The majority (76%) of ATRTs in the current study likewise had demonstrable losses of material from chromosome 22, including the SMARCB1 locus by FISH analysis.
Similar to Biegel et al,14 approximately one quarter of our ATRTs harbored homozygous deletions of SMARCB1. It is interesting to note that all of these cases, as well as the majority of those showing hemizygous SMARCB1 deletions, had much larger regions of loss involving chromosome 22. A high proportion of cells in these cases showed loss of both SMARCB1 (22q11.2) and a subtelomeric target, PANX2 (22q13.3), indicative of monosomy 22. Apart from the observations by Burger et al22 in a study that describes frequent monosomy 22 in ATRTs using a chromosome 22 paint probe, little comment has been made on the frequency of monosomy 22 in these tumors detectable by modern molecular techniques, particularly FISH. In most cases, locus-specific probes targeting the 22q11.2 region were paired with probes to nearby genes such as NF2 and EWS,9,11,28 which did not allow assessment of the distant subtelomeric region of chromosome 22 as was the case in the current study. Likewise, LOH and sequencing studies limit their focus to only SMARCB1.11,13,24 It would appear then that our findings and those of Burger et al22 indicate that monosomy 22, as opposed to smaller deletions limited to SMARCB1, may account for the significant proportion of the structural losses involving chromosome 22 in ATRT.
We found that in children 3 years or older at diagnosis of ATRT, treatment with high-dose alkylator-based chemotherapy and RT resulted in outcomes that are consistently superior to infants; eight of nine patients were alive at a median of 2.2 years from diagnosis. Disease recurrence developed in four patients in this cohort. Three patients are currently free of disease at 0.6, 1.5, and 9.5 years from diagnosis after therapy with ICE chemotherapy and with or without RT, suggesting that this combination may be a potential useful salvage therapy for this patient cohort.
Prognosis for infants and young children younger than 3 years at diagnosis of ATRT remains dismal. Our observations and those of others indicate younger patients are more likely to have disseminated disease at diagnosis and tend to develop disease progression and/or recurrence with higher frequency and earlier in the course of therapy than older patients. Further, in contrast to the older children, recurrent and/or progressive ATRT in children 3 years or younger appears refractory to salvage therapy.
Objective comparison of treatment efficacy between the two patient groups was problematic because of the heterogeneity in chemotherapeutic regimens administered and variability in the use of up-front RT. Disease progression occurred in 18 patients (82%) in the younger patient group and was seen early in the course of treatment. In all cases, recurrent disease occurred at the same time patients were receiving chemotherapy. None of the 18 patients was receiving concomitant RT, and only one patient (patient 15) had undergone prior RT. Further attempts at curative therapy after disease progression in this group were uniformly unsuccessful; all patients experienced a rapidly fatal course irrespective of the treatment (or lack thereof) rendered. Review of the literature demonstrates similar survival rates.29-31
RT has been associated with prolonged survival of older children and adults with ATRT. Hence, it is not surprising that in the current series the only long-term survivors in the younger patient cohort received RT early in the course of their treatment. Of the 18 young children with recurrent disease in our study, 15 experienced recurrence within 26 weeks of diagnosis, similar to other previous reports of brief progression-free intervals averaging 5 months (median, 4 months).2,20,22,29,31 It is notable that the only two long-term survivors (> 24 months from diagnosis) in our younger patient cohort both received RT as part of their up-front therapy. Additionally, the only patient in the older cohort who died was not treated with up-front RT. This prompts one to consider the utility of early focal RT for infants and young children with ATRT as the majority of patients develop progressive disease early, within 20 to 24 weeks of diagnosis.
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Julia Cay Jones for her editorial assistance and Jana Freeman for data management. We also thank the physicians and nursing staff who provided outstanding clinical care for the patients in this study.
NOTES
Supported by the Cancer Center (CORE) Support Grant CA 21765 from the National Institutes of Health and by the American Lebanese Syrian Associated Charities (ALSAC).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Kleihues P, Louis DN, Scheithauer BW, et al: The WHO classification of tumors of the nervous system. J Neuropathol Exp Neurol 61:215-225, 2002
2. Packer RJ, Beigel JA, Blaney S, et al: Atypical teratoid/rhabdoid tumor of the central nervous system: Report on workshop. J Pediatr Hematol Oncol 24:337-342, 2002
3. Ho DM-T, Hsu C-Y, Wong T-T, et al: Atypical teratoid/rhabdoid tumor of the central nervous system: A comparative study with primitive neuroectodermal tumor/medulloblastoma. Acta Neuropathol 99:482-488, 2000
4. Hilden JM, Meerbaum S, Burger P, et al: Central nervous system atypical teratoid/rhabdoid tumor: Results of therapy in children enrolled in a registry. J Clin Oncol 22:2877-2884, 2004
5. Lutterbach J, Liegibel J, Koch D, et al: Atypical teratoid/rhabdoid tumors in adult patients: Case report and review of the literature. J Neurooncol 52:49-56, 2001
6. Arrazola J, Pedrosa I, Mendez R, et al: Primary malignant rhabdoid tumour of the brain in an adult. Neuroradiology 42:363-367, 2000
7. Pimentel J, Silva R, Pimentel T: Primary malignant rhabdoid tumors of the central nervous system: Considerations about two cases of adulthood presentation. J Neurooncol 61:121-126, 2003
8. Berrak SG, zek MM, Canpolat C, et al: Association between DNA content and tumor suppressor gene expression and aggressiveness of atypical teratoid/rhabdoid tumors. Childs Nerv Syst 18:485-491, 2002
9. Fuller CE, Pfeifer J, Humphrey P, et al: Chromosome 22q dosage in composite extrarenal rhabdoid tumors: Clonal evolution or a phenotypic mimic Hum Pathol 32:1102-1108, 2001
10. Biegel JA, Allen CS, Kawasaki K, et al: Narrowing the critical region for a rhabdoid tumor locus in 22q11. Genes Chromosomes Cancer 16:94-105, 1996
11. Zhang Z-K, Davies KP, Allen J, et al: Cell cycle arrest and repression of cyclin D1 transcription by INI1/hSNF5. Molec Cell Biol 22:5975-5988, 2002
12. Biegel JA, Fan LT, Zhang F, et al: Alterations of the hSNF5/INI1 gene in central nervous system atypical teratoid rhabdoid tumors and renal and extrarenal rhabdoid tumors. Clin Cancer Res 8:3461-3467, 2002
13. Rousseau-Merck M-F, Versteege I, Legrand I, et al: hSNF5/INI1 Inactivation is mainly associated with homozygous deletions and mitotic recombinations in rhabdoid tumors. Cancer Res 59:3152-3156, 1999
14. Biegel JA, Fogelgren B, Wainwright LM, et al: Germline INI1 mutation in a patient with a central nervous system atypical teratoid/rhabdoid tumor. Genes Chromosom Cancer 20:31-37, 2002
15. Biegel JA, Zhou J-Y, Rorke LB, et al: Germ-line and acquired mutations of INI1 in atypical teratoid/rhabdoid tumors. Cancer Res 59:74-79, 1999
16. Dang T, Vassilyadi M, Michaud J, et al: Atypical teratoid/rhabdoid tumors. Childs Nerv Syst 19:244-248, 2003
17. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 54:457-81, 1958
18. Peto R, Pike MC, Armitage P, et al: I. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II. Analysis and examples. Br J Cancer 35:1-39, 1977
19. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163-170, 1966
20. Behring B, Brück W, Goebel HH, et al: Immunohistochemistry of primary central nervous system malignant rhabdoid tumors: Report of five cases and review of the literature. Acta Neuropathol 91:578-586, 1996
21. Merchant TE, Kun LE, Krasin MJ, Wallace D, et al: A multi-institutional prospective trial of reduced-dose craniospinal irradiation (23.4 Gy) followed by conformal posterior fossa (36 Gy) and primary site irradiation (55.8 Gy) and dose intensive chemotherapy for average-risk medulloblastoma. I J Radiat Oncol Biol Phys 57:194-195, 2000 (suppl, abstr)
22. Burger PC, Yu IT, Tihan T, 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 22:1083-1092, 1998
23. Fisher BJ, Siddiqui J, Macdonald D, et al: Malignant rhabdoid tumor of brain: An aggressive clinical entity. Can J Neurol Sci 23:257-263, 1996
24. Biegel JA, Fogelgren B, Zhou J-Y, et al: Mutations in the INI1 rhabdoid tumor suppressor gene in medulloblastomas and primitive neuroectodermal tumors of the central nervous system. Clin Cancer Res 6:2759-2763, 2000
25. Roberts CW, Orkin SH: The SWI/SNF complex–chromatin and cancer. Nat Rev Cancer 4:133-142, 2004
26. Rorke LB, Packer R, Biegel J: Central nervous system atypical teratoid/rhabdoid tumors of infancy and childhood. J Neurooncol 24:21-28, 1995
27. Betz BL, Strobeck MW, Reismann DN, et al: Re-expression of hSNF5/INI1/BAF47 in pediatric tumor cells leading to G1 arrest associated with induction of p16ink4a and activation of RB. Oncogene 21:5193-5203, 2002
28. Bruch LA, Hill DA, Cai DX, et al: A role for fluorescence in situ hybridization detection of chromosome 22q dosage in atypical teratoid/rhabdoid tumors from medulloblastoma/central primitive neuroectodermal tumors. Hum Pathol 32:156-162, 2001
29. Bambakidis NC, Robinson S, Cohen M, et al: Atypical teratoid/rhabdoid tumor of the central nervous system: Clinical, radiographic and pathologic features. Pediatr Neurosurg 37:64-70, 2002
30. Hilden JM, Watterson J, Longee DC, et al: Central nervous system atypical teratoid/rhabdoid tumor: Response to intensive therapy and review of the literature. J Neurooncol 40:265-275, 1998
31. Reinhardt D, Behnke-Mursch J, Weiss E, et al: Rhabdoid tumors of the central nervous system. Childs Nerv Syst 16:228-234, 2000(Tanya M. Tekautz, Christi)
Texas Children's Hospital, Houston TX
ABSTRACT
PATIENTS AND METHODS: Primary tumor samples from patients diagnosed with ATRT at SJCRH between July 1984 and June 2003 were identified. Pathology review included histologic, immunohistochemical analysis, and fluorescence in situ hybridization for SMARCB1 (also known as hSNF5/INI1) deletion. Clinical records of patients with pathologic confirmation of ATRT were reviewed.
RESULTS: Thirty-seven patients were diagnosed with ATRT at SJCRH during the 19-year study interval. Six patients were excluded from this clinical review based on pathologic or clinical criteria. Of the remaining 31 patients, 22 were younger than 3 years. Posterior fossa primary lesions and metastatic disease at diagnosis were more common in younger patients with ATRT. All patients underwent surgical resection; 30 received subsequent chemotherapy. The majority of patients aged 3 years or older received postoperative craniospinal radiation. Two-year event-free (EFS) and overall survival (OS) of children aged 3 years or older (EFS, 78% + 14%; OS, 89% ± 11%) were significantly better than those for younger patients (EFS, 11% ± 6%; OS, 17% ± 8%); EFS, P = .009 and OS, P = .0001. No other clinical characteristics were predictive of survival. Three of four patients 3 years or older with progressive disease were successfully rescued with ifosfamide, carboplatin, and etoposide therapy.
CONCLUSION: Children presenting with ATRT before the age of 3 years have a dismal prognosis. ATRT presenting in older patients can be cured using a combination of radiation and high-dose alkylating therapy. Older patients with relapsed ATRT can have salvage treatment using ICE chemotherapy.
INTRODUCTION
The diagnosis of ATRT relies predominantly on morphologic and immunohistochemical criteria (IHC).8 Chromosome 22q11 abnormalities are frequently seen in ATRT and have proved valuable for diagnostic purposes. Chromosome 22q11.2 harbors the putative tumor suppressor gene SMARCB1 (also known as hSNF5/INI1), a component of an SWI/SNF ATP-dependent chromatin-remodeling complex that likely controls the expression of certain target genes involved in cell cycle regulation.9-11 Seventy-five percent of ATRTs contain homozygous deletions of SMARCB1, or loss of one allele and mutation of the other copy of the gene.12-15 Published series of ATRT report an overall 2-year survival rate of less than 15% for children aged younger than 3 years at diagnosis. Therapeutic approaches to treatment of ATRT have included a variety of postoperative chemotherapy regimens with or without radiation therapy (RT).2,3,16 Although RT appears most effective when administered early in the treatment of ATRT,2 the unacceptable sequelae associated with cranial irradiation in infants and young children precludes the use of this modality in most patients with this disease.2 For the first time we present the clinical characteristics and survival correlates from our institutional series of patients with newly diagnosed ATRT and compare the infants and young children (< 3 years) with older children (≥ 3 years).
PATIENTS AND METHODS
Clinical Review
Once patients with the diagnosis of ATRT were identified, a detailed review of the clinical data was performed to ascertain the presenting features, degree of surgical resection, chemotherapy regimen, RT dose, and volume.
Pathology, Immunohistochemistry, and Fluorescence In Situ Hybridization
All available tumor material from patients receiving an institutional diagnosis of ATRT between July 1984 and June 2003 was reviewed by a single neuropathologist (C.F.). Histopathologic review included morphologic assessment of ATRT characteristics according to WHO criteria and immunohistochemical (IHC) analysis. IHC was performed using a panel of six antibodies including epithelial membrane antigen (EMA), smooth muscle actin (SMA), vimentin, glial fibrillary acidic protein (GFAP), keratin, and synaptophysin. The specific immunohistochemical stain(s) employed for a given tumor varied and was dependent on the histologic appearance of the tumor and amount of tissue available.
Tissue microarrays (TMA) were constructed from formalin-fixed, paraffin-embedded archival tissue blocks using a tissue arrayer (Beecher Instruments, Silver Spring, MD). Two to 4 tissue cores (1.0-mm diameter) from histologically representative areas of tumor were included from each case. Five micrometer-thick sections were then cut from the TMA and mounted on poly-L-lysine-coated slides for dual-color fluorescence in situ hybridization (FISH) experiments as previously reported.9 Bacterial artificial chromosome (BAC)–derived probes targeting SMARCB1 (22q11.23; contiguous gene sequence of CTD-2034E7 and CTD-2355C2, Invitrogen, Huntsville, AL) and PANX2 (control for chromosome 22 ploidy-22q13.3; RPCI3-402G11, Children's Hospital Oakland Research Institute, Oakland, CA) were labeled in rhodamine and fluorescein, respectively, and diluted 1:50 in DenHyb hybridization buffer (Insitus Laboratories, Albuquerque, NM) for dual-target hybridization. The probe and target were simultaneously denatured at 90°C for 13 minutes, followed by overnight hybridization at 37°C in a humidified chamber. DAPI (1.0 μg/mL; Insitus Laboratories) was used as a nuclear counterstain, and the sections were viewed under a Nikon E800 fluorescent microscope with appropriate filters (Nikon Instruments, Melville, NY).
Sections showing sufficient hybridization efficiency were considered informative, and 100 to 200 intact nonoverlapping nuclei were scored for the number of fluorescent signals by two reviewers (C.F., J.D.). Cutoffs for abnormalities and/or deletions were based on counts from non-neoplastic control specimens (normal brain from autopsy cases) for each probe. Interpretation of deletion required more than 37% of tumor nuclei containing 1 (hemizygous) or zero (homozygous) red SMARCB1 signals (mean + 3 SDs in controls); all cases showing homozygous deletion retained demonstrable SMARCB1 signals within normal tissue components (endothelial cells). As nuclei with more than two signals were rarely seen in non-neoplastic controls, polysomies (gains) were arbitrarily defined as more than 5% nuclei containing three or more signals.
Images were captured using a high-resolution black and white COHU CCD camera and the CytoVision basic workstation (Applied Imaging, Santa Clara, CA). A Z-stack motor allowed for sequential DAPI (one level), FITC (16 levels), and rhodamine (16 levels) filter settings to be captured, and the resulting images were reconstructed with blue, green, and red pseudocolors.
Statistical Analysis
Overall survival (OS) was measured from the date of initial diagnosis of ATRT to the date of death or last contact. Event-free survival (EFS) was measured from the date of diagnosis to the first date of progressive disease for those who progressed, to the date of death for those patients who died without progressive disease or to the date of last contact for children who survived in the absence of treatment failure. Distributions of EFS and OS were estimated using the method of Kaplan and Meier.17 Associated SEs were calculated as suggested by Peto et al.18 Differences in OS and EFS distributions in relation to age at diagnosis were assessed by exact log-rank tests.19 In comparison with survival estimates in relation to other risk factors, stratification was based on age at diagnosis; stratified exact log-rank tests were also used. Only two-sided P values are reported. Exact tests were performed using the StatXact software package (Cytel Software Corp, Cambridge, MA).
RESULTS
Pathology and Molecular Review
There was a broad range of histopathologic and immunohistochemical features present in the tumors from this patient cohort. In addition to a primitive neuroectodermal component composed of generally patternless sheets of round blue cells, tumor samples from all patients harbored variable numbers of "rhabdoid cells" with abundant cytoplasm often containing eosinophilic filamentous inclusions, eccentrically situated nuclei with open chromatin, and large nucleoli. IHC analysis of 31 ATRTs demonstrated strong cytoplasmic positivity with EMA, the majority additionally showing polyphenotypic immunoprofiles consistent with ATRT including positivity for SMA, vimentin, GFAP, keratin, and synaptophysin, in decreasing order of frequency (Table 1). 8,9,20
Twenty-one tumor samples from 20 patients were available for FISH analysis (Table 1). Sixteen cases (76%) contained deletion of SMARCB1. Homozygous SMARCB1 deletion was detected in five cases (24%); the average percentage of cells showing homozygous deletion in these five samples was 84% (range, 67% to 92%). All five tumors that displayed homozygous deletion also displayed evidence of concomitant monosomy 22, with only a single PANX2 signal per cell (average percentage of cells, 51%; range, 34% to 80%). These data suggest that in these cases homozygous deletion occurred as a result of monosomy 22 and associated interstitial deletion of the second SMARCB1 allele. Eight cases (38%) showed deletion of a single copy of SMARCB1/hemizygous deletion (average percentage of cells, 52%; range, 39% to 93%). A significant proportion of these cases showed a larger region of loss on chromosome 22; up to 63% (range, 50% to 71%) of INI1-deleted cells displaying single red and green signals (monosomy) and a lesser percentage had only hemizygous loss of INI1 (single red INI1 and two green PANX2 signals per cell). Monosomy 22 alone was the predominant alteration in three cases (14%) with the majority of tumor cells involved (average, 82%; range, 76% to 86%). Only five cases (24%) had no demonstrable alterations of chromosome 22 detected by the current FISH assay. In one patient (patient 5), FISH analysis on tumor obtained at diagnosis did not detect any INI1 abnormality. The child subsequently developed recurrent disease; evaluation of the recurrent tumor demonstrated an INI1 hemizygous deletion by FISH.
ATRT Patient Cohort: Demographic/ Clinical Features
Twenty-two patients (71%) were younger than 3 years at diagnosis with a median age of 1 year (range, 0.2 to 2.7 years). Sixteen patients younger than 3 years old at diagnoses were male; 16 were white, four were African-American, and two were Hispanic. The median age at diagnosis of patients 3 years and older was 3.9 years (range, 3.3 to 7.4 years) and included three males; six were white, two were African Americans, and one was Hispanic.
Disease extent was evaluated by computed tomography and/or magnetic resonance (MR) imaging of the brain and spine after tumor resection and, Tc99 bone scintigraphy. CSF cytology was examined in all cases.
RT was delivered either to the entire neuraxis or locally. The dose of craniospinal irradiation (CSI) administered was determined by the patient's age, postoperative residual disease, and the presence of metastatic disease. Patients 3 years or older with less than 1.5 cm2 postoperative residual disease and no evidence of neuraxis dissemination received 2,340 cGy to the neuraxis whereas all other patients 3 years or older were treated to a neuraxis dose of 3,600 cGy. The primary site of disease or tumor bed was treated to a dose of 5,580 cGy using image-guided techniques.21 Patients who were treated with local-only RT had image-guided delivery of their treatment to cumulative doses selected by the treating physicians.
Primary tumors originated from the posterior fossa (n = 14, 45%), the supratentorial region (n = 16; 52%), and spine (n = 1). Posterior fossa was the most common primary tumor site among patients younger than 3 years (n = 12; 55%). In contrast, supratentorial primary tumors were predominant in the older patients (n = 7; 78%); only two (22%) tumors originated in the posterior fossa. Metastatic disease was present in six patients (27%) at diagnosis (M1 = 2, M2 = 2, M3 = 2); one patient (patient 17) was found to have an asymptomatic renal rhabdoid tumor. All patients with initial metastatic disease were younger than 3 years at diagnosis (Table 2).
Treatment for all patients included surgical resection of the tumor (Table 2). Of the 31 patients, 21 (68%) with ATRT underwent gross total or near total tumor resection (GTR/NTR); 14 of these patients were younger than 3 years. Ten patients (32%) had subtotal resections (STR), eight patients were younger than 3 years at diagnosis. Treatment administered after surgical excision varied depending on patient age (Table 3). All patients younger than 3 years were treated with chemotherapy, three also received RT or CSI. Seven of the nine patients 3 years or older were treated with RT and chemotherapy as outlined in Tables 3 and 4.
Therapeutic Response and Survival
No chemotherapy-related deaths occurred. Of the 22 patients younger than 3 years at diagnosis, 18 experienced recurrent or progressive disease at a median of 0.4 years from diagnosis (range, 0.1 to 0.7 years). Recurrence was local in 13 (72%), distant in 3 (17%), and both local and distant in 2 (11%). Thirteen of the 18 patients with relapsed disease were treated with chemotherapy (n = 4), RT (n = 4), or a combination of chemotherapy and RT (n = 5; Table 3). The five patients who did not receive further treatment experienced a median survival of 0.3 years (range, 0 to 0.3 years). Salvage treatment consisting of chemotherapy only was administered to four patients. Treatment included alkylator-based chemotherapy (cyclophosphamide/cisplatin/vincristine/etoposide; carboplatin/ifosfamide/etopside (ICE); high-dose cyclophosphamide and topotecan followed by autologous hematopoietic stem-cell reconstitution; and oxaliplatin/topotecan). The median survival was 0.3 years (range, 0.3 to 0.5 years).
Radiation only was employed in the treatment of four patients with recurrent disease, three of four patients had CSI (3,300 to 3,960 cGy), one received local RT (2,100 cGy). The median survival of these patients was 0.4 years (range, 0 to 0.8 years). Finally, five patients were treated with RT and chemotherapy after recurrence was detected. Radiation was administered before the chemotherapy in four of these children. Two were treated with CSI (2,500 and 3,960 cGy) and three received local RT (5,040 to 5,580 cGy). Chemotherapy in this group included two patients who received high-dose chemotherapy with autologous hematopoietic stem cell reconstitution (one of whom was treated with intrathecal methotrexate) in combination with an alkylator-based chemotherapy regimen. One patient received cisplatin and etoposide, and one patient was treated with a topoisomerase I inhibitor. The median survival in this group was 0.6 years (range, 0.3 to 1.4 years). All patients younger than 3 years with recurrent disease died within 1.5 years of progression with a median survival of 0.4 years (range, 0.06 to1.4 years).
Four of the 22 patients in the younger cohort did not develop recurrent or progressive disease. One patient (patient 16) died of respiratory failure (secondary to complications of surgery at the time of tumor resection) 2.5 months from diagnosis without evidence of progressive disease. The remaining three patients are alive without recurrent disease. One patient (patient 18) who received chemotherapy only electively terminated treatment at 6 months from diagnosis and is without evidence of disease 10 months from diagnosis. Two patients were treated with surgical resection, chemotherapy, and RT; they are alive 4.8 and 5.8 years from diagnosis (patients 7 and 6, respectively).
In contrast, of the nine patients 3 years of age or older at diagnosis, five remain free of disease at a median of 2.2 years (range, 0.8 to 4.1 years) from diagnosis. Four have experienced disease recurrence at a median of 1.2 years from diagnosis (range, 0.2 to 3.9 years; Table 3); recurrence was local in two and both local and distant in two. Two of these patients (patients 23 and 24) did not receive any RT as part of their initial therapy. Three of the four patients who experienced disease recurrence are alive (two with no evidence of disease; one alive with disease) at 0.6, 1.5, and 9.5 years after five to eight cycles of ifosphamide, carboplatin, etoposide, and RT in two of the three (local, 1; CSI, 1).
EFS and OS estimates for the entire cohort of 31 patients at 2 years were 31% ± 9% and 40% ± 10%, respectively. The median length of follow-up for the group younger than 3 years was 0.7 years (range, 0.2 to 5.8 years), and their 2-year EFS and OS estimates were 11% ± 6% and 17% ± 8%, respectively (Figs. 1A and B). For the group 3 years and older at diagnosis, the median length of follow-up was 2.8 years (range, 0.5 to 9.8 years); the 2-year EFS estimate was 78% ± 14%, and the 2-year OS estimate was 89% ± 11% (Figs. 1A and 1B). Given the striking association between age and EFS (P = .009; Fig 1A) and OS (P = .0001; Fig 1B), further analyses were stratified based on age at diagnosis. Primary tumor location was not predictive of EFS (P = .41). Similarly, EFS was not correlated with M-stage at diagnosis (P = .95), extent of the initial tumor resection (P = .096), or abnormalities in chromosome 22 (P = .82). Separate analysis of the two age groups, however, suggested an association between the extent of surgical resection (GTR/NTR v STR) and OS estimates (P = 0.053; data not shown).
Survival estimates for those patients initially treated with chemotherapy alone were compared with those for patients treated with chemotherapy and RT. In this analysis, we did not consider the chemotherapeutic agents used or doses administered, the extent and dose of RT, or the sequence in which these treatments were given. For the 21 patients treated with chemotherapy, the 1-year EFS estimate was 0%, the 1-year OS estimate was 42% ± 11%, and a 2-year OS estimate was 12% ± 7% (Figs. 2A and 2B). In contrast, the 2-year EFS and OS estimates for the 10 patients treated with chemotherapy and RT (three from the younger group and seven from the older group) were both 90% ± 10% (Figs. 2A and 2B). The correlation between treatment administered and EFS (P = .0003) and OS (P = .007) in our patient cohort with RT associated with improved survival. Therapy received was dependant on age, disease location, and disease extent, and thus the survival differences observed are likely attributable to multiple confounding factors. It is notable, however, only two long-term survivors (> 24 months from diagnosis) in children younger than 3 years at diagnosis from the current cohort both received RT as part of their up-front therapy, before the development of progressive disease.
DISCUSSION
Molecular and cytogenetic analysis initially implicated abnormalities of 22q11.2 in the pathogenesis of both ATRT and malignant rhabdoid tumor (MRT)12,13 and subsequently, the SMARCB1 gene was identified in this region.10 As part of the SWI/SNF chromatin remodeling complex, SMARCB1 functions as a DNA-binding protein, facilitating SWI/SNF-mediated cotransactivation of genes involved in cell cycle regulation.13,25,26 Homozygous deletions or truncation mutations of the SMARCB1gene abrogate SWI/SNF-imposed cell-cycle arrest11,13 and have been identified in up to 75% of ATRTs evaluated.9,13,27,28 The majority (76%) of ATRTs in the current study likewise had demonstrable losses of material from chromosome 22, including the SMARCB1 locus by FISH analysis.
Similar to Biegel et al,14 approximately one quarter of our ATRTs harbored homozygous deletions of SMARCB1. It is interesting to note that all of these cases, as well as the majority of those showing hemizygous SMARCB1 deletions, had much larger regions of loss involving chromosome 22. A high proportion of cells in these cases showed loss of both SMARCB1 (22q11.2) and a subtelomeric target, PANX2 (22q13.3), indicative of monosomy 22. Apart from the observations by Burger et al22 in a study that describes frequent monosomy 22 in ATRTs using a chromosome 22 paint probe, little comment has been made on the frequency of monosomy 22 in these tumors detectable by modern molecular techniques, particularly FISH. In most cases, locus-specific probes targeting the 22q11.2 region were paired with probes to nearby genes such as NF2 and EWS,9,11,28 which did not allow assessment of the distant subtelomeric region of chromosome 22 as was the case in the current study. Likewise, LOH and sequencing studies limit their focus to only SMARCB1.11,13,24 It would appear then that our findings and those of Burger et al22 indicate that monosomy 22, as opposed to smaller deletions limited to SMARCB1, may account for the significant proportion of the structural losses involving chromosome 22 in ATRT.
We found that in children 3 years or older at diagnosis of ATRT, treatment with high-dose alkylator-based chemotherapy and RT resulted in outcomes that are consistently superior to infants; eight of nine patients were alive at a median of 2.2 years from diagnosis. Disease recurrence developed in four patients in this cohort. Three patients are currently free of disease at 0.6, 1.5, and 9.5 years from diagnosis after therapy with ICE chemotherapy and with or without RT, suggesting that this combination may be a potential useful salvage therapy for this patient cohort.
Prognosis for infants and young children younger than 3 years at diagnosis of ATRT remains dismal. Our observations and those of others indicate younger patients are more likely to have disseminated disease at diagnosis and tend to develop disease progression and/or recurrence with higher frequency and earlier in the course of therapy than older patients. Further, in contrast to the older children, recurrent and/or progressive ATRT in children 3 years or younger appears refractory to salvage therapy.
Objective comparison of treatment efficacy between the two patient groups was problematic because of the heterogeneity in chemotherapeutic regimens administered and variability in the use of up-front RT. Disease progression occurred in 18 patients (82%) in the younger patient group and was seen early in the course of treatment. In all cases, recurrent disease occurred at the same time patients were receiving chemotherapy. None of the 18 patients was receiving concomitant RT, and only one patient (patient 15) had undergone prior RT. Further attempts at curative therapy after disease progression in this group were uniformly unsuccessful; all patients experienced a rapidly fatal course irrespective of the treatment (or lack thereof) rendered. Review of the literature demonstrates similar survival rates.29-31
RT has been associated with prolonged survival of older children and adults with ATRT. Hence, it is not surprising that in the current series the only long-term survivors in the younger patient cohort received RT early in the course of their treatment. Of the 18 young children with recurrent disease in our study, 15 experienced recurrence within 26 weeks of diagnosis, similar to other previous reports of brief progression-free intervals averaging 5 months (median, 4 months).2,20,22,29,31 It is notable that the only two long-term survivors (> 24 months from diagnosis) in our younger patient cohort both received RT as part of their up-front therapy. Additionally, the only patient in the older cohort who died was not treated with up-front RT. This prompts one to consider the utility of early focal RT for infants and young children with ATRT as the majority of patients develop progressive disease early, within 20 to 24 weeks of diagnosis.
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Julia Cay Jones for her editorial assistance and Jana Freeman for data management. We also thank the physicians and nursing staff who provided outstanding clinical care for the patients in this study.
NOTES
Supported by the Cancer Center (CORE) Support Grant CA 21765 from the National Institutes of Health and by the American Lebanese Syrian Associated Charities (ALSAC).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Kleihues P, Louis DN, Scheithauer BW, et al: The WHO classification of tumors of the nervous system. J Neuropathol Exp Neurol 61:215-225, 2002
2. Packer RJ, Beigel JA, Blaney S, et al: Atypical teratoid/rhabdoid tumor of the central nervous system: Report on workshop. J Pediatr Hematol Oncol 24:337-342, 2002
3. Ho DM-T, Hsu C-Y, Wong T-T, et al: Atypical teratoid/rhabdoid tumor of the central nervous system: A comparative study with primitive neuroectodermal tumor/medulloblastoma. Acta Neuropathol 99:482-488, 2000
4. Hilden JM, Meerbaum S, Burger P, et al: Central nervous system atypical teratoid/rhabdoid tumor: Results of therapy in children enrolled in a registry. J Clin Oncol 22:2877-2884, 2004
5. Lutterbach J, Liegibel J, Koch D, et al: Atypical teratoid/rhabdoid tumors in adult patients: Case report and review of the literature. J Neurooncol 52:49-56, 2001
6. Arrazola J, Pedrosa I, Mendez R, et al: Primary malignant rhabdoid tumour of the brain in an adult. Neuroradiology 42:363-367, 2000
7. Pimentel J, Silva R, Pimentel T: Primary malignant rhabdoid tumors of the central nervous system: Considerations about two cases of adulthood presentation. J Neurooncol 61:121-126, 2003
8. Berrak SG, zek MM, Canpolat C, et al: Association between DNA content and tumor suppressor gene expression and aggressiveness of atypical teratoid/rhabdoid tumors. Childs Nerv Syst 18:485-491, 2002
9. Fuller CE, Pfeifer J, Humphrey P, et al: Chromosome 22q dosage in composite extrarenal rhabdoid tumors: Clonal evolution or a phenotypic mimic Hum Pathol 32:1102-1108, 2001
10. Biegel JA, Allen CS, Kawasaki K, et al: Narrowing the critical region for a rhabdoid tumor locus in 22q11. Genes Chromosomes Cancer 16:94-105, 1996
11. Zhang Z-K, Davies KP, Allen J, et al: Cell cycle arrest and repression of cyclin D1 transcription by INI1/hSNF5. Molec Cell Biol 22:5975-5988, 2002
12. Biegel JA, Fan LT, Zhang F, et al: Alterations of the hSNF5/INI1 gene in central nervous system atypical teratoid rhabdoid tumors and renal and extrarenal rhabdoid tumors. Clin Cancer Res 8:3461-3467, 2002
13. Rousseau-Merck M-F, Versteege I, Legrand I, et al: hSNF5/INI1 Inactivation is mainly associated with homozygous deletions and mitotic recombinations in rhabdoid tumors. Cancer Res 59:3152-3156, 1999
14. Biegel JA, Fogelgren B, Wainwright LM, et al: Germline INI1 mutation in a patient with a central nervous system atypical teratoid/rhabdoid tumor. Genes Chromosom Cancer 20:31-37, 2002
15. Biegel JA, Zhou J-Y, Rorke LB, et al: Germ-line and acquired mutations of INI1 in atypical teratoid/rhabdoid tumors. Cancer Res 59:74-79, 1999
16. Dang T, Vassilyadi M, Michaud J, et al: Atypical teratoid/rhabdoid tumors. Childs Nerv Syst 19:244-248, 2003
17. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 54:457-81, 1958
18. Peto R, Pike MC, Armitage P, et al: I. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II. Analysis and examples. Br J Cancer 35:1-39, 1977
19. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163-170, 1966
20. Behring B, Brück W, Goebel HH, et al: Immunohistochemistry of primary central nervous system malignant rhabdoid tumors: Report of five cases and review of the literature. Acta Neuropathol 91:578-586, 1996
21. Merchant TE, Kun LE, Krasin MJ, Wallace D, et al: A multi-institutional prospective trial of reduced-dose craniospinal irradiation (23.4 Gy) followed by conformal posterior fossa (36 Gy) and primary site irradiation (55.8 Gy) and dose intensive chemotherapy for average-risk medulloblastoma. I J Radiat Oncol Biol Phys 57:194-195, 2000 (suppl, abstr)
22. Burger PC, Yu IT, Tihan T, 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 22:1083-1092, 1998
23. Fisher BJ, Siddiqui J, Macdonald D, et al: Malignant rhabdoid tumor of brain: An aggressive clinical entity. Can J Neurol Sci 23:257-263, 1996
24. Biegel JA, Fogelgren B, Zhou J-Y, et al: Mutations in the INI1 rhabdoid tumor suppressor gene in medulloblastomas and primitive neuroectodermal tumors of the central nervous system. Clin Cancer Res 6:2759-2763, 2000
25. Roberts CW, Orkin SH: The SWI/SNF complex–chromatin and cancer. Nat Rev Cancer 4:133-142, 2004
26. Rorke LB, Packer R, Biegel J: Central nervous system atypical teratoid/rhabdoid tumors of infancy and childhood. J Neurooncol 24:21-28, 1995
27. Betz BL, Strobeck MW, Reismann DN, et al: Re-expression of hSNF5/INI1/BAF47 in pediatric tumor cells leading to G1 arrest associated with induction of p16ink4a and activation of RB. Oncogene 21:5193-5203, 2002
28. Bruch LA, Hill DA, Cai DX, et al: A role for fluorescence in situ hybridization detection of chromosome 22q dosage in atypical teratoid/rhabdoid tumors from medulloblastoma/central primitive neuroectodermal tumors. Hum Pathol 32:156-162, 2001
29. Bambakidis NC, Robinson S, Cohen M, et al: Atypical teratoid/rhabdoid tumor of the central nervous system: Clinical, radiographic and pathologic features. Pediatr Neurosurg 37:64-70, 2002
30. Hilden JM, Watterson J, Longee DC, et al: Central nervous system atypical teratoid/rhabdoid tumor: Response to intensive therapy and review of the literature. J Neurooncol 40:265-275, 1998
31. Reinhardt D, Behnke-Mursch J, Weiss E, et al: Rhabdoid tumors of the central nervous system. Childs Nerv Syst 16:228-234, 2000(Tanya M. Tekautz, Christi)