Malignant Rhabdoid Tumor

Updated: Oct 09, 2018
  • Author: James I Geller, MD; Chief Editor: Max J Coppes, MD, PhD, MBA  more...
  • Print



Malignant rhabdoid tumor (MRT) is one of the most aggressive and lethal malignancies in pediatric oncology. Malignant rhabdoid tumor was initially described in 1978 as a rhabdomyosarcomatoid variant of a Wilms tumor because of its occurrence in the kidney and because of the resemblance of its cells to rhabdomyoblasts. The absence of muscular differentiation led Haas and colleagues to coin the term rhabdoid tumor of the kidney in 1981. [1]

Although renal malignant rhabdoid tumor was historically included in treatment protocols of the National Wilms Tumor Study (NWTS) Group, this tumor is now recognized as an entity separate from a Wilms tumor. In contrast to a Wilms tumor, a MRT of the kidney is characterized by the early onset of local and distant metastases and resistance to chemotherapy. Whereas the overall survival rate for Wilms tumors exceeds 85%, the survival rate for renal MRTs is only 20-25%.

Since MRT of the kidney was originally described, MRTs have been reported in practically every location in the body, including the brain, liver, soft tissues, lung, skin, and heart. This article focuses on renal and extrarenal MRTs that arise outside the CNS.

Molecular genetics

Cytogenetic, fluorescence in situ hybridization (FISH), and loss-of-heterozygosity (LOH) studies have revealed that MRTs frequently contain deletions at chromosome locus 22q11.1. Positional cloning efforts revealed that this locus contains the SWI/SNF related, matrix-associated, actin-dependent regulator of chromatin, subfamily B, member 1 (SMARCB1) gene, also known as human sucrose non-fermenting gene number 5 (hSNF5), integrase interactor 1 (INI1), or 47-Kd Brg1/Bam-associated factor (BAF47). [2]  SMARCB1 encodes a member of the human SWI/SNF complex.

Combined analyses including FISH, coding sequence analysis, high-density single nucleotide polymorphism-based oligonucleotide arrays, and multiplex ligation-dependent probe amplification enable the identification of biallelic, inactivating perturbations of SMARCB1 in nearly all MRTs, consistent with the 2-hit model of tumor formation. [3] Thus, SMARCB1 is presumed to function as a classic tumor suppressor and the primary gene responsible for MRT development.

Homozygous inactivation of SMARCB1 in mice demonstrates embryonic lethality, whereas heterozygous SMARCB1 mice demonstrate a normal phenotype at birth, with 20% developing sarcomas at a median age of 1 year. Similar to human MRTs, murine tumors in these mice acquire a second hit to the SMARCB1 locus. All mice harboring a conditional biallelic inactivation of SMARCB1 develop cancer with a median onset of 11 weeks, revealing one of the most aggressive cancer predisposition genotype-phenotype correlations known.

Unexpectedly, despite an aggressive clinical pattern of behavior, MRTs are generally diploid and genomically stable, without recurrent gene amplifications or deletions. Therefore, the mechanism by which SMARCB1 perturbation leads to aggressive neoplasia likely relates to its role in epigenetic modification. The SWI/SNF complex acts in an adenosine triphosphate (ATP)–dependent manner to remodel chromatin, which regulates gene transcription and DNA repair. Whole-exome sequencing studies of primary MRTs have shown that biallelic mutations or copy number alterations of SMARCB1 seem to be both necessary and sufficient to cause cancer. [4]

The distribution of SMARCB1 and chromosome 22-inactiving mutations, deletions, and copy number loss of heterozygosity (CN-LOH) in 200 sporadic AT/RTs, renal MRTs, and extrarenal MRTs is shown in Table 1 below. In the majority of tumors (43%), there is a mutation in one allele, and the second copy of the gene is lost as a result of a structural deletion in 22q11.2, monosomy 22, or an acquired CN LOH event. Compound heterozygous mutations are infrequent in these patients (4%). Partial deletions and duplications are detected in approximately 15% of tumors. Homozygous deletions of exons 1-9 of SMARCB1 are present in approximately 40% of rhabdoid tumors overall, although there is an unequal distribution with respect to anatomic location. Approximately 25% of AT/RTs, 40% of renal MRTs, and 70% of extrarenal MRTs have homozygous deletions of the entire locus. [4]

Table 1. Acquired SMARCB1 Alterations in 200 Sporadic Rhabdoid Tumors (Adapted from Geller JI, Roth JJ, Biegel JA. Biology and Treatment of Rhabdoid Tumor. Crit Rev Oncog. 2015. 20 [3-4]:199-216) (Open Table in a new window)

Allele 1 Alteration

Allele 2 Alteration



Partial Gene Deletion/Duplication

Whole Gene Deletion



8 (4%)

1 (0.50%)

58 (29%)

27 (13.5%)

94 (47%)

Partial Gene Deletion/Duplication


5 (2.5%)

14 (7%)

11 (5.3%)

30 (15%)

Whole Gene Deletion



76 (38%)


76 (38%)


8 (4%)

6 (3%)

148 (74%)

38 (19%)

200 (100%)

Although SMARCB1 is the predominant gene altered in MRTs, approximately 2-3% retain expression of the SMARCB1 protein on immunohistochemistry and do not display inactivating mutations in the gene. [4] A small number of families and patients with MRT have been shown to have germline or somatic mutations in SMARCA4, which is the primary ATPase in the SWI/SNF complex. [5, 6]




The histogenetic origin of malignant rhabdoid tumor (MRT) remains obscure. Rhabdoid tumor cells are polyphenotypic, with an immunostaining pattern that shows evidence of mesenchymal, epithelial, and neural differentiation. Polyantigenic expression suggests that MRTs arises from a pluripotent cell capable of differentiating along several lines.

Considerable debate has been focused on whether extrarenal MRTs are the same as renal MRTs. The recognition that CNS AT/RTs have deletions of the SMARCB1 gene indicates that MRTs of the kidney and brain are closely related entities. This observation is not surprising because rhabdoid tumors at both locations possess similar histologic, clinical, and demographic features. Moreover, 10-15% of patients with non-CNS MRTs have synchronous or metachronous brain tumors, many of which are second primary AT/RTs.

More convincingly, the same germline mutations (outside of frameshift mutations) predispose carriers to atypical teratoid/rhabdoid tumor (AT/RT), renal MRTs, and to a lesser extent extrarenal MRTs. The majority of extrarenal MRTs are sporadic and arise as a consequence of homozygous loss of SMARCB1. The most frequent second hit in patients with a germline mutation is a large 22q deletion or monosomy 22, or a copy number loss of heterozygosity (CN-LOH) generating event. [4]

Reports to date have demonstrated that SMARCB1 loss can promote cell cycle progression resulting from upregulation of targets of the p16INK4a-Rb-E2F pathway, primarily including CyclinD1 as well as several cyclin-dependent kinases (CDK). Rb family loss has been shown to increase MRT tumorigenesis and progression, whereas ablation of CyclinD1 abrogates MRT evolution in mouse models. Similarly, tumor development in SMARCB1-deficient mice is greatly accelerated in the absence of functional p53 protein. These findings suggest a cooperative effect between SMARCB1 and the pRB, CyclinD1, and Tp53 pathways. The loss of SMARCB1 is postulated to result in a global failure of the repressive H3K27 trimethylation mark present on bivalently modified histones, mediated by the polycomb complex 2, resulting in widespread epigenetic modifications and leading to arrested development and abnormal proliferation. Two members of the polycomb complex 2, CBX6 and EZH2, are upregulated in MRT. [7] Aurora-A-kinase is also expressed in high levels in MRT and is repressible with SMARCB1 reintroduction into rhabdoid tumor cells via transcriptional down-regulation. SMARCB1 loss also leads to increased expression of GLI1, supporting a role in the biology of the sonic hedgehog pathway. [8] Bromodomain containing protein 9 (BRD9) is also a subunit of the SWI/SNF complex that is involved in epigenetic mechanisms such as regulation of transcription, chromatin remodeling and histone modification. [9] These findings suggest possible therapeutic targets for MRT. For CNS MRT (atypical teratoid/rhabdoid tumor (AT/RT)), biologic heterogeneity has been discovered. There are three distinct molecular subgroups of AT/RT tumors, termed ATRT-TYR, ARTR-SHH, and ATRT-MYC, that are associated with differences in demographics, tumor location, and type of SMARCB1 mutations. [10] The ATRT-TRY group is composed of mostly infratentorial tumors with broad SMARCB1 deletions and overexpression of melanosomal genes. The ATRT-SHH group is associated with both supra- and infratentorial lesions, has focal SMARCB1 aberrations, and shows overexpression of SHH pathways. The ATRT-MYC group is comprised of mostly supratentorial tumors with focal SMARCB1 deletions and over-expression of the MYC and HOX cluster. [10] The development of these molecular subgroups with associated regulatory networks and pathways will help to develop more effective, subgroup specific treatment options.

Similarly, comprehensive genomic analysis was performed on non-CNS MRTs through the Therapeutically Applicable Research to Generate Effective Targets (TARGET) initiative. [11] This analysis showed evidence for epigenetic reprogramming of HOX genes with loss of H3K27me3 marks, as well as dysregulated expression of genes involved in neural crest development and of oncogenes and tumor suppressor genes. Although SMARCB1 alterations were shown to be the dominant genetic driver in MRTs, the effects of that loss on transcriptional regulation were not uniform across all cases. Conversely, the spectrum of tumors characterized by mutations in the SMARCB1 gene has also been expanded beyond tumors with a rhabdoid histologic phenotype to include hereditary schwannomas, extraskeletal myxoid chondrosarcoma, [12] proximal-type epithelioid sarcoma, epithelioid malignant peripheral nerve sheath tumor, renal medullary carcinoma, [13] chordoma, and pediatric undifferentiated sarcoma lacking rhabdoid features. [14] Inactivation of SMARCB1 has also been identified in the small cell undifferentiated variant of hepatoblastoma, which may suggest that such tumors are better characterized as hepatic rhabdoid tumors. [15, 16] Whether extrarenal MRTs have the same histogenetic origin as that of their renal counterparts is unclear. Although some extrarenal MRTs are considered to be undifferentiated sarcomas or carcinomas with rhabdoid features, others represent true rhabdoid tumors because they have documented SMARCB1 mutations. [17] To date, the only tumor outside of MRT to demonstrate biallelic inactivation of SMARCA4, consistent with a cancer-predisposing germline mutation and second somatic alteration, is small cell carcinoma of the ovary, hypercalcemic type (SCCOHT). [18] Based on the early age at presentation and presence of rhabdoid-appearing cells on histology, it has been proposed that SCCOHT represents another type of extrarenal MRT. [19]

The Children's Oncology Group (COG) has initiated an effort to prospectively screen all types of MRT for SMARCB1 mutations and protein expression, which should improve the classification and prognostication of tumors with rhabdoid features. As molecular-based targeted therapies emerge, the distinction between true and pseudorhabdoid tumors may prove to have important therapeutic implications.

For details about the gross and histologic features of MRTs, see Histologic Findings below.




United States

Malignant rhabdoid tumor (MRT) is a rare tumor. According to registration data from NWTS 1-5, MRT of the kidney accounts for only 158 (1.6%) of 10,031 registrants with childhood renal tumors. Likewise, only 26 (0.9%) of 3000 participants in the Intergroup Rhabdomyosarcoma Studies I-III had tumors consistent with MRT. Of the first 4,000 patients enrolled on the COG study AREN03B2, renal and extrarenal MRTs accounted for 3.71% combined. [20] About 15 cases of extrarenal, non-CNS MRTs are diagnosed each year in the North America. MRTs in infants (age 0-12 months) account for 18% of renal tumors, 14% of soft tissue tumors, and 9% of liver tumors. [21]


The incidence of MRT in most countries has not been reported. Between 1984 and 1999, approximately 6 patients per year diagnosed with MRT were enrolled onto various national registries or protocols in Germany. [22] Between 1993 and 2005, a total of 207 renal MRT K patients were enrolled on SIOP renal tumor treatment study group protocols. [23]


The overall survival rate for patients with MRT enrolled in NWTS 1-5 was 23.2%. [24]

MRT is a rapidly progressive tumor, with most deaths occurring within 12 months of presentation. The most common sites of metastasis at presentation are the lungs, abdominal lymph nodes, liver, brain, and less commonly bone (1.4%).

A young age at diagnosis is strongly associated with an adverse outcome. Four-year event-free survival rates according to age at diagnosis were 8.8% for patients aged 0-5 months, 17.2% for patients aged 6-11 months, 28.6% for patients aged 12-23 months, and 41.1% for patients aged 24 months or older (p < 0.0001). [24]

High-stage (stage III/IV) disease is also correlated with an adverse outcome (p=0.014); most patients present with stage III or IV disease.

The survival of patients with MRT in NWTS was as follows:

  • Stage I - 15 patients (33.3%)

  • Stage II - 25 patients (46.9%)

  • Stage III - 58 patients (21.8%)

  • Stage IV - 41 patients (8.4%)

  • Stage V - 3 patients (0%)

In a smaller study of 70 patients with MRT and AT/RT from Germany, metastatic disease at diagnosis maintained prognostic value, although age did not. [22] Additional preliminary data suggest that patients with germline mutations of SMARCB1 likely manifest disease at an earlier age, with a high risk of progression and inferior prognosis.

Survival outcomes remained poor for those patients with higher stage disease treated on the most recent COG study AREN0321 (regimen UH1), a more intensive chemotherapy regimen than previously used in NWTSG trials. The 4-year EFS and OS for the entire cohort were both 35.9%. [25]  

The survival of patients with MRT in AREN0321 was as follows:

  • Stage I - 2 patients (100%)
  • Stage II - 5 patients (100%)
  • Stage III - 23 patients (26.1%)
  • Stage IV - 9 patients (11.1%)
  • Stage V - 0 patients

Such data suggest that intensive therapy may benefit rare low stage MRT patients (stage 1 and 2), but novel therapy is necessary for stage 3 and 4 patients.


MRT has no apparent racial predilection.


MRT occurs slightly more frequently in male individuals than in female individuals, with male-to-female ratio of 1.37:1. [24]


The median age at presentation is 10.6 months, with a mean age of 15 months. Most patients are younger than 2 years. MRT has been reported in children older than this and in adults, but whether older patients have a biologically distinct subtype of MRT is unclear.