Clinical manifestations

Presentation and course

As is the case with other brain tumors, patients with atypical teratoid/rhabdoid tumor present with signs and symptoms that reflect the location of the tumor. Approximately half of atypical teratoid/rhabdoid tumors arise in the posterior fossa (40). Young patients with posterior fossa tumors usually present with symptoms related to hydrocephalus, namely early morning headaches, vomiting, and lethargy. They may also develop ataxia or regression of motor skills. Cranial nerve palsies, particularly VI and VII, are not uncommon. Because young children have open sutures and fairly pliable skulls, a rapidly enlarging head size should alert the practitioner to evaluate the patient for a brain mass as this may be the only symptom at presentation. Patients with supratentorial tumors can present with signs and symptoms of hydrocephalus, enlarging head size, hemiparesis, seizures, or visual changes depending of the location of the tumors. Because atypical teratoid/rhabdoid tumor is a highly malignant tumor, patients typically have a fairly short history of progressive symptoms measured in days to weeks. It is unusual for symptoms to have been present for more than a few months. No data support that clinical characteristics can separate atypical teratoid/rhabdoid tumors from other malignant brain tumors.

Imaging characteristics are also helpful but nonspecific for atypical teratoid/rhabdoid tumor areas of necrosis, hemorrhage, and associated perilesional edema. On T1-weighted imaging, the tumor mass is typically isointense with frequent hyperintense foci secondary to intratumoral hemorrhage. These tumors are highly cellular; therefore, they are hypointense in T2-weighted sequences and demonstrate diffusion restriction in most cases (50). MR spectroscopy findings were similar to other embryonal tumors with marked elevation of choline and low or absent N-acetyl-aspartate and creatine.

On imaging, supratentorial atypical teratoid/rhabdoid tumors are characterized by a thick, wavy, and heterogeneously enhancing wall; hence, imaging is useful when differentiating atypical teratoid/rhabdoid tumors from other entities (02). Additionally, Koral and colleagues reported that tumors located within the cerebellopontine angle (CPA), particularly those presenting with intertumoral hemorrhage, are more likely to correspond to atypical teratoid/rhabdoid tumors than medulloblastomas (33).

Although clinical and radiographic findings can suggest atypical teratoid/rhabdoid tumor, the diagnosis is made pathologically. Loss of immune expression of the SMARCB1 (protein: INI-1) is highly specific for the diagnosis of atypical teratoid/rhabdoid tumor (30). Epithelial membrane antigen (EMA), vimentin, and smooth-muscle actin (SMA) immunopositivity are also commonly seen. A minority of cases (1% to 2%) present with retained INI-1 expression and Brahma-related gene 1 (BRG-1) negativity due to pathogenic mutations in SMARCA4 (23).

The diagnosis of atypical teratoid/rhabdoid tumor should be considered as part of the differential diagnosis of young children, especially when the child is less than 3 years of age presenting with a malignant brain tumor.

Prognosis and complications

The prognosis for patients with atypical teratoid/rhabdoid tumors has improved over the last decades after increased recognition as a distinct entity and the development of atypical teratoid/rhabdoid tumor-specific approaches.

Nevertheless, the prognosis of these patients remains guarded, particularly in the younger patients.

Clinical vignette

A 13-year-old previously healthy female presented to the emergency department with a 1-month history of progressive worsening headaches, and over the week before presentation, the headaches were associated with early morning emesis. She had no vision changes, focal neurologic deficits, or gait disturbances. She was initially treated symptomatically with pain medication; due to the lack of symptomatic improvement, brain imaging was performed. MRI of the brain demonstrated a large solid and cystic lesion located in the right frontal lobe measuring 5.4 cm x 6.1 cm. The mass was partially hemorrhagic and demonstrated diffusion restriction on diffusion-weighted imaging of the spine and did not demonstrate evidence of disseminated disease.

The patient underwent a gross total tumor resection; she tolerated the procedure well and had no neurologic deficits postoperatively.

Histopathological examination demonstrated a markedly hypercellular neoplasm with nuclear enlargement and prominent nucleoli. Immunohistochemistry showed negativity for GFAP, olig2, S100, and synaptophysin. INI-1 expression was lost in the tumor cells, supporting the diagnosis of atypical teratoid/rhabdoid tumor. Methylation profiling classified this tumor as an ATRT-MYC subgroup.

Staging workup was negative for disseminated disease, and renal ultrasound showed normal kidneys without a renal mass.

The patient was formally enrolled and treated in a therapeutic trial that included craniospinal proton irradiation (CSI) 23.4 Gy and tumor bed boost to a total dose of 54 Gy; she received consolidation chemotherapy with five cycles of intensive chemotherapy with cyclophosphamide, cisplatin, vincristine, and an Aurora kinase inhibitor (Alisertib). Finally, she received six cycles of maintenance of alisertib. Currently, the patient is off therapy without evidence of disease and will continue under close clinical and radiological surveillance.

Biological basis

Etiology and pathogenesis

Pathology. Atypical teratoid/rhabdoid tumor is recognized as a distinct clinical entity because of its pathologic and genetic characteristics. Histologically, atypical teratoid/rhabdoid tumor contains sheets of rhabdoid cells against a background of primitive neuroectodermal cells, mesenchymal cells, or epithelial cells. Some tumors are composed entirely of rhabdoid cells, whereas others show a combination of rhabdoid cells and areas resembling primitive neuroectodermal tumor or medulloblastoma. Immunohistochemical features help to identify the disease but vary depending on the cellular composition of the tumor. Rhabdoid cells express vimentin and epithelial membrane antigen, neurofilament, glial fibrillary acidic protein, and keratin. The primitive neuroectodermal cells may express neurofilament protein, glial fibrillary acidic protein, or desmin. These rapidly growing tumors can have K-67 labeling indices of 50% to 100% (32). Finally, the cardinal immunohistochemical finding for diagnosis is the loss of expression of INI-1 or BRG-1 (30).

Pathogenesis and molecular subgrouping. Atypical teratoid/rhabdoid tumors are molecularly characterized by biallelic inactivation of the SWItch/Sucrose Non-Fermentable (SWI/SNF)-related, matrix-associated, actin-dependent regulator of chromatin, subfamily b, member 1 (SMARCB1) or, in rare cases subfamily a, member 4 (SMARCA4).

Epigenetic dysregulation has been postulated as the main mechanism of oncogenesis in this entity. The loss of a member of the chromatin remodeling SNF/SWF complex, INI-1 or BRG-1, increases the expression and activity of polycomb repressive complexes (PRCs), thereby activating the histone methyl transferase enhancer of zeste homolog 2 (EZH2) downstream oncogenic pathways (36; 27).

Studies postulate that different residual INI-1-deficient SWI/SNF complexes are associated with the activation of different lineage-related pathways, accounting for the oncogenic phenotypes observed in atypical teratoid/rhabdoid tumors (38; 49). This discovery led to the molecular classification of atypical teratoid/rhabdoid tumor, which established three molecular subgroups: ATRT-SHH, ATRT-TYR, and ATRT-MYC (26; 47; 28).

The prognostic significance of these subgroups remains unclear (20; 41), and additional studies are underway to better understand the subgroup-specific therapeutic susceptibilities.

Epidemiology

Atypical teratoid/rhabdoid tumor was initially thought to be extraordinarily rare. With increased clinical awareness of this tumor and the utility of the immunohistochemistry assay for SMARCB1 (INI-1), it is now estimated that atypical teratoid/rhabdoid tumor accounts for at least 3% of brain tumors in children. All published series report a peak incidence in very young children, less than 3 years of age. Based on prior Children’s Cancer Group, Pediatric Oncology Group, and Pediatric Brain Tumor Consortium studies. Atypical teratoid/rhabdoid tumors account for approximately 20% of brain tumors in children less than 3 years of age (35; 52; 35; 24). In an Austrian study of 311 newly diagnosed tumors, atypical teratoid/rhabdoid tumor was the sixth most common entity (6.1%) (52). The number of reported patients with atypical teratoid/rhabdoid tumor has increased 5.5-fold in an analysis of the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) data from 1992 to 2008, likely reflecting greater awareness and improved diagnostic techniques. The Central Brain Tumor Registry of the United States (CBTRUS) reported an age-adjusted incidence rate of 0.9 per million children between the ages of 0 to 19 years from 2012 to 2016 (39). Although it is much less common in older children and adults, atypical teratoid/rhabdoid tumor has been reported with increasing frequency in these populations as well.

Prevention

There is no known means to prevent the development of an atypical teratoid rhabdoid tumor. In patients harboring germline SMARCB1 (RTPS1) or SMARCA4 (RTPS2) mutations, surveillance imaging with whole-body MRIs (WBMRI) and brain/spine MRIs have been recommended, with the objective to diagnose tumors at a stage on which complete resection is still possible (21).

Differential diagnosis

The rhabdoid tumor predisposition syndromes (RTPS1 and 2) are rare genetic conditions characterized by germline truncating pathogenic variants in SMARCB1 (RTPS1) or SMARCA4 (RTPS2) genes. Those affected by the condition have an increased risk of developing rhabdoid tumors (intra and extracranially) (44; 34; 07; 15).

Rhabdoid tumor predisposition syndrome is inherited in an autosomal dominant manner. Although most cases occur de novo, in rare cases, individuals inherit a pathogenic variant from an unaffected parent.

It is estimated that 20% to 35% of patients diagnosed with atypical teratoid/rhabdoid tumors across all age groups have an associated deleterious germline alteration (04). Nevertheless, there is a wide variation among different studies. The published ACNS0333 study only identified a pathogenic germline mutation in 17% of patients in their cohort (41). In contrast, Fossey and colleagues reported a higher proportion of patients (83.3%) diagnosed with atypical teratoid/rhabdoid tumor in the first year of life, carrying a pathogenic mutation in SMARCB1 (18).

Although the prognostic significance of germline alterations in patients with atypical teratoid/rhabdoid tumor remains to be elucidated, it provides useful information for counseling and surveillance purposes (19; 21). The recommendation is to offer germline testing and genetic counseling to all patients diagnosed with atypical teratoid/rhabdoid tumors.

Confusing conditions

Atypical teratoid/rhabdoid tumor may be difficult to distinguish from other embryonal tumors, particularly medulloblastoma. Before atypical teratoid/rhabdoid tumor was widely recognized, many were misdiagnosed as such (42; 08). Other malignant nervous system tumors should also be considered in the differential diagnosis, including embryonal tumors with multilayer rosettes (ETMRs), pineoblastomas, choroid plexus carcinomas, and ependymoma. High-grade glial neoplasms can occur in young children but are extremely rare.

Diagnostic workup

Patients with suspected atypical teratoid/rhabdoid tumor should undergo complete MRI imaging of the brain and spine prior to surgery unless emergent surgery precludes this being done. This is to quantify the radiographic extent of disease at diagnosis and is necessary for conformal radiation planning. Surgical resection is indicated for nearly all patients with atypical teratoid/rhabdoid tumor; this is for both diagnosis and treatment. The tumor should undergo pathologic and molecular genetic testing. Methylation profiling should be considered if adequate tumor is available. A lumbar puncture should be performed as part of the staging evaluation to assess for cerebrospinal fluid dissemination. The cerebrospinal fluid evaluation should be completed at least 14 days after surgery to reduce the probability of false-positive results (37).

Patients with new diagnosis of atypical teratoid/rhabdoid tumor should undergo a renal ultrasound to rule out the possibility of synchronous tumors.

Management

Treatment of atypical teratoid/rhabdoid tumor has often paralleled that of infant medulloblastoma or other embryonal tumors. Early trials by the Pediatric Oncology Group POG) 9233/4 and Children’s Cancer Group (CCG) 9921 treated very young children diagnosed with various embryonal brain tumors, including atypical teratoid/rhabdoid tumors with multi-agent chemotherapy, resulting in poor outcomes (22; 45). Multimodal therapeutic approaches have improved the survival of these patients (25; 10; 48; 35; 03). The standard of care includes a combination of maximal safe resection, chemotherapy, and radiation therapy.

Surgery. Surgical resection is an important component in the management of atypical teratoid/rhabdoid tumors. However, due to the invasive nature of these tumors, gross total resection (GTR) is only possible in approximately 30% to 68% of cases (46; 10; 14; 35). Some studies have suggested a survival advantage in patients undergoing a gross total resection (10); however, other studies failed to demonstrate a prognostic effect (46; 01; 43).

Contemporary data from the ACNS0333 trial demonstrated that the extent of surgical resection did not impact survival when part of an intensified multimodal approach including high-dose chemotherapy (HDC) and radiotherapy (RT) (41). Likewise, the European Rhabdoid Registry (EU-RHAB) reported the outcomes of 143 patients uniformly treated using either radiotherapy or high-dose chemotherapy (20); akin to the North American results, complete surgical resection was not identified as a prognostic factor in association with other therapeutic modalities. These data suggest a maximal safe surgical resection should continue to be sought after in patients with atypical teratoid/rhabdoid tumor.

Chemotherapy. Given the young age of most patients with atypical teratoid/rhabdoid tumor, chemotherapy has been the main form of postsurgical adjuvant therapy. Varieties of chemotherapeutic approaches have been utilized with varying success. Many patients have also been treated with radiation, adding further complexity to evaluating the efficacy of a given chemotherapy regimen. Outcomes for children treated with surgery and standard chemotherapy alone are extremely poor.

Legacy cooperative group “baby brain” protocols employed dose-intensified multi-agent chemotherapy to treat a variety of malignant brain tumors in very young children POG 9233/4 and CCG 9921 enrolled 63 patients with atypical teratoid/rhabdoid tumor and achieved a 24-months event-free survival (EFS) of 6.4% (22; 45). In both studies, however, early responses to chemotherapy were seen.

Chi and colleagues published the results of the first prospective trial using the backbone of the Intergroup Rhabdomyosarcoma Study protocol III (IRS-III) and modified it to include intrathecal chemotherapy (10). The study resulted in a 58% response rate before radiation therapy, suggesting a survival benefit using intensified multimodal therapy.

High-dose chemotherapy was adopted as a radiotherapy deferral/avoidance approach (25; 10; 16; 35). Hilden and colleagues reported their experience using high-dose chemotherapy in patients with atypical teratoid/rhabdoid tumor, demonstrating 46% of patients who received high-dose chemotherapy were survivors (25). Furthermore, Cohen and colleagues reported the outcomes of eight patients with atypical teratoid/rhabdoid tumor treated in the CCG99703 study, demonstrating a 5-year event-free survival of 37.5% (12).

The COG-ACNS0333 study incorporated high-dose methotrexate to the CCG99703 chemotherapy backbone followed by consolidation therapy with three tandem cycles of HDC with carboplatin and thiotepa conditioning with stem cell rescue and radiotherapy. Sixty-five evaluable patients had a 4-year event-free survival and overall survival (OS) of 37% and 43%, respectively (41).

The European Rhabdoid Registry (EU-RHAB) reported the outcomes of 143 patients uniformly treated using either radiotherapy or high-dose chemotherapy and reported a 5-year event-free survival and overall survival of 30.5% and 34.7%, respectively (20).

Radiation therapy appears to be an important component of therapy for atypical teratoid/rhabdoid tumor patients. Several studies have demonstrated a survival advantage in patients with atypical teratoid/rhabdoid tumor treated with radiotherapy. Tekautz and colleagues published the outcomes of nine children over the age of 3 years treated with craniospinal irradiation (CSI) in addition to high-dose chemotherapy as part of their initial treatment (46). A 3-year event-free survival of 78% was reported.

Buscariollo and colleagues identified 144 patients with atypical teratoid/rhabdoid tumor in the National Cancer Institute’s Surveillance, Epidemiology, and End Results Database (09). The study confirmed a robust association between radiation therapy use and survival.

ACNS0333 led to significantly improved survival and confirms that some patient with atypical teratoid/rhabdoid tumor can be cured of their disease with intensive multimodal regimens that include high-dose chemotherapy with peripheral blood stem cells and radiation.

Proton beam radiation may have an advantageous role in treating atypical teratoid/rhabdoid tumors. A series of very young patients treated with proton beam reported good disease control and increased sparing of healthy tissues, such as cerebrum, temporal lobe, cochlea, and hypothalamus (13).

Future directions. Approximately half of the patients will unfortunately succumb to the disease despite the use of multimodal approaches. The improved understanding of the biological underpinnings of atypical teratoid/rhabdoid tumor will result in the identification of new therapeutic targets.

Atypical teratoid/rhabdoid tumors are a disease of epigenetic dysregulation mediated by the loss of SMARCB1 (51). The defective polycomb repressive complex (PCR) leads to increased expression and activity of the histone methyl transferase enhancer of zeste homolog 2 (EZH2) (51; 38). In-vitro studies have demonstrated that cell lines derived from ATRT-SHH tumors are particularly sensitive to EZH2 inhibition (47); hence, a phase I trial for SMARCB1-deficient tumors is underway (NCT02601937).

Preclinical data suggest atypical teratoid/rhabdoid tumor may be a good candidate for pathway-specific therapies. Several preclinical studies have shown promising results. Analyses demonstrated upregulation of platelet-derived growth factor receptor B (PDGFRB) in ATRT-MYC and ATRT-TYR cell lines and xenografts. Dasatinib and nilotinib produced significant in vitro and in vivo cytotoxicity (47). Although further studies are required to validate these observations, the findings provide promising targets for future studies.

Furthermore, studies have improved our understanding of the immune microenvironment in ATRT. Chun and colleagues demonstrated ATRT-MYC tumors had significantly greater CD8+ T-cell infiltration than tumors in the TYR/SHH subgroups (11). Several clinical trials using immunotherapy for ATRT are ongoing.

The progress made toward understanding tumor biology will facilitate the development of novel, risk-stratified, and biologically informed therapeutic approaches.