SMARCB1-deficient cancers are a group of tumors that share a common genetic feature: loss of the SMARCB1 protein, a component of the cell’s gene-regulation machinery that normally acts as a tumor suppressor. When both copies of the SMARCB1 gene are knocked out, cells lose a critical brake on growth, giving rise to cancers that tend to be aggressive and difficult to treat. These tumors span a surprisingly wide range of tissue types and ages, from rare brain tumors in infants to soft-tissue sarcomas and kidney cancers in young adults, which makes the shared molecular defect all the more striking as a unifying thread.
What SMARCB1 Does in Healthy Cells
SMARCB1 (also known as INI1 or BAF47) is a core subunit of a large protein machine called the SWI/SNF chromatin-remodeling complex. This complex works like a molecular doorman: it physically shifts the packaging of DNA so that certain genes can be read by the cell or shut off when they should be silent. The SWI/SNF complex has been described as a prototype of an epigenetic regulator involved in tumor suppression, meaning it controls gene activity not by changing the DNA sequence itself but by managing how tightly DNA is wound around its protein spools.1PubMed. SWI/SNF Chromatin-remodeling Complex Status in SMARCB1/INI1-preserved Epithelioid Sarcoma When SMARCB1 is present and functioning, it helps the SWI/SNF complex keep growth-promoting genes in check and activate genes that push cells toward normal maturity. When SMARCB1 is lost, the complex can no longer do that job properly, and the cell’s gene-expression landscape shifts in ways that favor uncontrolled growth.
A key part of this shift involves a rival protein complex called PRC2 (Polycomb Repressive Complex 2). In a healthy cell, SWI/SNF and PRC2 act as opposing forces: SWI/SNF opens up regions of the genome, while PRC2 slaps silencing marks on them. Under normal conditions, SWI/SNF can push PRC2 off the promoters of tumor-suppressor genes, keeping those genes active. When SMARCB1 is lost and SWI/SNF function degrades, PRC2 goes unopposed, silencing tumor suppressors and enabling cancer-driving programs to take over.2PubMed Central. Overcoming Clinical Resistance to EZH2 Inhibition Using Rational Epigenetic Combination Therapy This antagonism between SWI/SNF and PRC2 is not just a textbook detail; it has become a direct target for therapy, as we will see.
The Cancers That Arise from SMARCB1 Loss
The prototypical SMARCB1-deficient tumor is the malignant rhabdoid tumor, first identified in the kidney but capable of arising in soft tissue and the brain. Rhabdoid tumors overwhelmingly strike very young children, and most follow an aggressive course with poor outcomes.3PubMed. SMARCB1-deficient Tumors of Childhood: A Practical Guide When rhabdoid tumors occur in the central nervous system, they are called atypical teratoid/rhabdoid tumors (ATRTs). ATRTs are rare but among the most aggressive pediatric brain tumors, and their genetic hallmark is biallelic inactivation of SMARCB1.4PubMed Central. Current Molecular and Clinical Landscape of ATRT – The Link to Future Therapies Together, these rhabdoid tumors account for a disproportionate share of cancer deaths in infants and toddlers relative to how uncommon they are.
In adults, the picture looks different but the molecular defect is the same. Epithelioid sarcoma, a soft-tissue cancer that typically appears in the hands, forearms, or trunk of young adults, loses SMARCB1 at very high rates. One study found that roughly 90% of epithelioid sarcomas tested negative for SMARCB1 protein by immunohistochemistry.5PubMed Central. Epithelioid sarcoma is associated with a high percentage of SMARCB1 deletions The loss rate can vary by subtype: proximal-type epithelioid sarcomas (which tend to arise in deeper tissues near the trunk) lose SMARCB1 expression in about 76% of cases, while conventional distal-type tumors show loss in about 93%.6PubMed. Infrequent SMARCB1/INI1 gene alteration in epithelioid sarcoma: a useful tool in distinguishing epithelioid sarcoma from malignant rhabdoid tumor Epithelioid sarcoma was, in fact, the first adult cancer to receive FDA approval for a therapy specifically targeting this SMARCB1-driven vulnerability.
Renal medullary carcinoma is another cancer defined by SMARCB1 loss, and it has an unusual demographic fingerprint: it almost exclusively develops in young people who carry sickle cell trait or sickle cell disease.7PubMed Central. SMARCB1 regulates the hypoxic stress response in sickle cell trait The connection to sickle cell trait appears to involve chronic low-grade oxygen stress in the kidney’s inner medulla, where sickling red blood cells create conditions that may predispose cells to losing SMARCB1 function. The tumor is always characterized by SMARCB1 loss and is driven by biallelic inactivation of the gene.8Nature Communications. SMARCB1 regulates a TFCP2L1-MYC transcriptional switch promoting renal medullary carcinoma transformation and ferroptosis resistance Renal medullary carcinoma is highly aggressive and often presents at an advanced stage, making early recognition critical.
Beyond these better-known entities, SMARCB1-deficient sinonasal carcinoma has been recognized as a distinct tumor type arising in the nasal cavity and paranasal sinuses. A pooled analysis of published cases found that outcomes differ by stage and sex, with patients presenting at later stages having considerably shorter survival.9PubMed Central. SMARCB1 (INI-1)-Deficient Sinonasal Carcinoma: A Systematic Review and Pooled Analysis of Treatment Outcomes More broadly, SMARCB1-deficient tumors in adults represent a diverse group with mixed phenotypes, which can make diagnosis challenging and prognosis often poor, with relapsing and refractory disease being common.10PubMed Central. SMARCB1/INI1-deficient tumors of adulthood
How These Cancers Are Diagnosed
The single most useful diagnostic tool across all SMARCB1-deficient cancers is immunohistochemistry for the SMARCB1 (INI1) protein. In a normal tissue sample, essentially every cell nucleus stains positive for SMARCB1. In a SMARCB1-deficient tumor, the cancer cells show complete loss of nuclear staining while the surrounding non-tumor cells still stain normally, providing a sharp visual contrast under the microscope. This staining pattern has become a valuable tool both for confirming and screening for SMARCB1-deficient tumors.11Advances in Anatomic Pathology. The Expanding Family of SMARCB1(INI1)-deficient Neoplasia: Implications of Phenotypic, Biological, and Molecular Heterogeneity
The practical importance of this stain cannot be overstated. Many SMARCB1-deficient tumors look similar to other cancers under the microscope. Rhabdoid tumors can mimic other small round blue-cell tumors of childhood. Epithelioid sarcoma can be confused with carcinoma or even benign conditions like granuloma. A study examining rhabdoid tumors across the brain, kidney, and soft tissue confirmed that neoplastic cells consistently failed to express SMARCB1 by immunohistochemistry, while all other tumor types tested retained expression.12PubMed. Absence of expression of SMARCB1/INI1 in malignant rhabdoid tumors of the central nervous system, kidneys and soft tissue: an immunohistochemical study with implications for diagnosis That binary staining result, present in normal tissue and absent in tumor cells, turns an otherwise ambiguous biopsy into a clear diagnosis. In many institutions, SMARCB1 immunohistochemistry is now routinely performed on any tumor that fits the clinical or morphological profile of a rhabdoid or SMARCB1-deficient cancer.
Molecular testing for SMARCB1 gene alterations (deletions, mutations, or both) can confirm the immunohistochemistry result and is especially useful in cases where staining is equivocal. Biallelic inactivation, meaning both copies of the gene are knocked out, is the hallmark finding. In malignant rhabdoid tumors, molecular analysis confirms biallelic SMARCB1 inactivation in the vast majority of cases.3PubMed. SMARCB1-deficient Tumors of Childhood: A Practical Guide
When SMARCB1 Loss Runs in Families
Most SMARCB1-deficient cancers arise from mutations acquired during a person’s lifetime, but in a subset of cases the first hit to SMARCB1 is inherited. This gives rise to the rhabdoid tumor predisposition syndromes (RTPS1 for inherited SMARCB1 variants, and the related RTPS2 for inherited SMARCA4 variants, another SWI/SNF subunit). Children born with one defective copy of SMARCB1 already carry half the genetic setup for tumor formation; all it takes is loss of the remaining copy in a susceptible cell, and a rhabdoid tumor can develop.13PubMed Central. Current recommendations for clinical surveillance and genetic testing in rhabdoid tumor predisposition: a report from the SIOPE Host Genome Working Group
The penetrance of RTPS1 early in life is high, meaning that a large proportion of children carrying a pathogenic SMARCB1 variant will develop tumors, often before age two. Survival in these cases has historically been poor. The predisposing variants are characteristically truncating, meaning they cut the gene’s message short and effectively eliminate the protein product. This distinguishes RTPS from other conditions linked to SMARCB1, such as Coffin-Siris syndrome, where different types of variants (often missense rather than truncating) cause developmental abnormalities without the same cancer risk.13PubMed Central. Current recommendations for clinical surveillance and genetic testing in rhabdoid tumor predisposition: a report from the SIOPE Host Genome Working Group For families with a known SMARCB1 germline variant, clinical surveillance recommendations now exist to catch tumors early, though the rarity of the condition means evidence-based screening protocols are still evolving.
Molecular Subtypes Within SMARCB1-Deficient Brain Tumors
Not all ATRTs behave the same way, and researchers have found that SMARCB1-mutated ATRTs can be divided into at least three molecular subgroups based on their DNA methylation profiles: ATRT-TYR, ATRT-SHH, and ATRT-MYC.14PubMed Central. Atypical teratoid/rhabdoid tumors (ATRTs) with SMARCA4 mutation are molecularly distinct from SMARCB1-deficient cases These subgroups are not just academic labels. They correspond to different locations in the brain, different ages at diagnosis, and potentially different responses to therapy.
In a large-scale analysis of over 160 ATRTs, the subgroups showed distinct patterns: one cluster arose predominantly in the cerebral hemispheres above the tentorium, while another concentrated in the cerebellum and brainstem below.15Cancer Cell. Integrated (epi)-Genomic Analyses Identify Subgroup-Specific Therapeutic Targets in CNS Rhabdoid Tumors The MYC subgroup of ATRTs shares features with rhabdoid tumors arising outside the brain, including widespread loss of DNA methylation marks and activation of developmental genes normally expressed in mesenchymal (connective tissue) cells. These similarities suggest that the MYC subgroup and extra-cranial rhabdoid tumors may arise from overlapping cell types or developmental pathways, setting them apart from ATRTs that express more neural features.16PubMed Central. Identification and Analyses of Extra-Cranial and Cranial Rhabdoid Tumor Molecular Subgroups Reveal Tumors with Cytotoxic T Cell Infiltration Understanding these subgroups matters because treatments that work for one ATRT subtype might not work for another, and clinical trials are increasingly designed with this molecular classification in mind.
Targeting the PRC2 Vulnerability with EZH2 Inhibitors
The antagonistic relationship between SWI/SNF and PRC2 suggested an obvious therapeutic strategy: if SMARCB1-deficient cells become excessively dependent on PRC2 to silence tumor suppressors, then blocking PRC2’s enzymatic engine, a protein called EZH2, might reactivate those silenced genes and stop the cancer. This reasoning led to the development of tazemetostat, an EZH2 inhibitor that became the first drug approved specifically for epithelioid sarcoma in adults.
Results in pediatric SMARCB1-deficient tumors have been more sobering. A trial run through the NCI-COG Pediatric MATCH program tested tazemetostat in children with refractory tumors harboring SMARCB1 or related alterations. The objective response rate was only about 5%, falling short of the primary efficacy endpoint. However, a quarter of patients experienced prolonged stable disease lasting six months or longer, suggesting the drug may slow progression even when it does not shrink tumors dramatically.17PubMed Central. Tazemetostat for tumors harboring SMARCB1/SMARCA4 or EZH2 alterations: results from NCI-COG pediatric MATCH APEC1621C Among those who experienced prolonged disease stabilization, all had tumors with SMARCB1 loss specifically, including one epithelioid sarcoma patient who stayed on treatment for 26 cycles.18JNCI: Journal of the National Cancer Institute. Tazemetostat for tumors harboring SMARCB1/SMARCA4 or EZH2 alterations: results from NCI-COG pediatric MATCH APEC1621C
The limited response rates have pushed researchers to investigate why resistance develops and how to overcome it. Functional genomics studies of SMARCB1-deficient tumor models have revealed that resistance to tazemetostat often converges on the RB1/E2F pathway, essentially finding an alternative way to keep dividing despite EZH2 being blocked. The tumors acquire mutations that decouple two things tazemetostat is supposed to do simultaneously: force the cancer cells to mature and stop them from dividing. When resistance mutations allow cells to escape the growth arrest while still being pushed toward differentiation, the drug loses its killing power.2PubMed Central. Overcoming Clinical Resistance to EZH2 Inhibition Using Rational Epigenetic Combination Therapy This work has identified potential biomarkers that could predict which patients will respond and has opened the door to combination strategies designed to close off escape routes.
Immunotherapy for Tumors with Simple Genomes
One of the more surprising findings in SMARCB1-deficient cancer research is that some of these tumors respond to immune checkpoint inhibitors despite having very simple genomes with low tumor mutational burden. Conventional wisdom holds that cancers need many mutations to generate the abnormal proteins (neoantigens) that attract immune-cell attention. Rhabdoid tumors and renal medullary carcinomas have remarkably few mutations beyond the SMARCB1 loss itself, yet preclinical data and clinical case reports suggest they can be immunogenic.19PubMed Central. Immunotherapy for SMARCB1-Deficient Sarcomas: Current Evidence and Future Developments
The mechanism behind this apparent immunogenicity is still being worked out, but one hypothesis is that the epigenetic disruption caused by SMARCB1 loss reactivates normally silent genomic elements, creating signals that the immune system can recognize even without a high mutation count. Case reports have documented responses to PD-1 inhibitors in SMARCB1-deficient tumors, including at least one instance in a rectal carcinoma that had low mutational burden and intact mismatch repair, both features that would usually predict a poor response to checkpoint blockade.20PubMed Central. Response to PD-1 inhibitor in SMARCB1‑deficient undifferentiated rectal carcinoma with low TMB, proficient MMR and BRAF V600E mutation: a case report and literature review This has led some researchers to propose that SMARCB1 loss itself could serve as a predictive biomarker for immune checkpoint blockade, though prospective trials are needed before that idea becomes standard practice.
The molecular subgroup classification mentioned earlier may play a role here too. Among rhabdoid tumors, certain subgroups show evidence of infiltration by cytotoxic T cells, the immune cells responsible for killing cancer, while others do not.16PubMed Central. Identification and Analyses of Extra-Cranial and Cranial Rhabdoid Tumor Molecular Subgroups Reveal Tumors with Cytotoxic T Cell Infiltration If validated, this could mean immunotherapy has the most promise in specific molecular subtypes rather than across all SMARCB1-deficient cancers equally.
Synthetic Lethality and the Search for New Drug Targets
When a cancer cell has already lost one important gene, it often becomes unusually dependent on a backup pathway to survive. If you knock out that backup, the cancer cell dies while normal cells, which still have both pathways intact, carry on unharmed. This concept is called synthetic lethality, and it has become a major focus in SMARCB1-deficient cancer research. The identification of synthetic lethal targets based on the vulnerabilities created by SWI/SNF complex deficiency is seen as a contribution to precision medicine for these tumors.21PubMed Central. Synthetic lethal therapy based on targeting the vulnerability of SWI/SNF chromatin remodeling complex-deficient cancers
High-throughput screening technologies have accelerated the search for synthetic lethal partners of SMARCB1 loss. Large-scale genetic screens using CRISPR and related tools have uncovered numerous genes whose loss is tolerated in normal cells but lethal in cells lacking SWI/SNF function.22PubMed Central. Epigenetic synthetic lethality approaches in cancer therapy Some of these targets are other epigenetic regulators, some are cell-cycle machinery, and some involve metabolic pathways that SMARCB1-deficient cells lean on more heavily than their normal counterparts. The EZH2 inhibitor approach already described is itself an example of exploiting a synthetic lethal interaction, since blocking EZH2 is only toxic to cells that have already lost SWI/SNF opposition. The field is now looking beyond EZH2 toward combinations and alternative targets that could help patients whose tumors resist EZH2 inhibition alone.
Preclinical Models and the Road Ahead
Studying SMARCB1-deficient cancers in the lab has historically been difficult because the tumors are rare and the cells do not always grow well in culture. Newer approaches have started to change that. One team used CRISPR gene editing to knock out SMARCB1 and TP53 in human induced pluripotent stem cells, then briefly guided them toward a neural fate to create cells that resemble ATRT. This model allowed them to screen drugs in a system that more faithfully mirrors the human disease, and initial testing identified several compounds that reduced the viability of the ATRT-like cells.23Bioactive Materials. Modeling human brain rhabdoid tumor by inactivating tumor suppressor genes in induced pluripotent stem cells
Better models matter because SMARCB1-deficient cancers are too rare for large randomized trials to be conducted quickly. Each tumor type within this family numbers in the low hundreds of cases annually worldwide, so progress depends on laboratory systems that accurately predict which drugs will work in patients. The convergence of improved preclinical models, molecular subtyping that can stratify patients for clinical trials, and a growing understanding of resistance mechanisms means the therapeutic landscape for SMARCB1-deficient cancers is more active today than at any point since the gene was first linked to cancer in the mid-1990s.24Cancer Research. SMARCB1/INI1 Tumor Suppressor Gene Is Frequently Inactivated in Epithelioid Sarcomas For patients and families affected by these diseases, the question is no longer whether targeted therapies can be built around SMARCB1 loss but whether they can be made potent and durable enough to change long-term outcomes.