The most well-characterized mechanism of EVI1 activation is through chromosomal rearrangements involving its locus at 3q26.2 (Table 1). The classic rearrangements are the pericentric inversion inv(3)(q21q26.2) and the balanced translocation t(3;3)(q21;q26.2). These abnormalities are found in approximately 2-3% of AML cases and are strongly associated with a unique clinical phenotype, often including normal or elevated platelet counts and multilineage dysplasia, particularly of the megakaryocytic lineage (20). Both inv(3) and t(3;3) are classified as high-risk aberrations conferring an extremely poor prognosis (20, 21).

The primary molecular consequence of these rearrangements is the repositioning of a distal hematopoietic enhancer of the GATA2 gene, located at 3q21, into close proximity with the EVI1 promoter at 3q26.2 (29). This GATA2 enhancer is highly active in hematopoietic stem and progenitor cells, and its hijacking by the EVI1 locus drives massive, lineage-inappropriate overexpression of EVI1 (2, 30). This mechanism exemplifies a phenomenon known as “enhancer hijacking,” where a potent regulatory element is aberrantly co-opted to drive the expression of a proto-oncogene (Figure 2). In addition to these classic 3q abnormalities, other, more cryptic or atypical rearrangements involving the 3q26 band have been identified that similarly result in EVI1 overexpression and are associated with a clinical course resembling that of classic inv(3) AML (31). The consistent outcome of these diverse rearrangements is the deregulation of EVI1, establishing it as the critical pathogenic driver associated with 3q26 abnormalities. This mechanism also often leads to haploinsufficiency of GATA2, further contributing to the hematopoietic defects observed in these patients.

A second major mechanism for EVI1 activation is the formation of chimeric fusion genes through chromosomal translocations that fuse the coding sequence of EVI1 with that of another gene. Besides the classical inv(3)/t(3;3), a number of other 3q26 rearrangements with poor treatment response have been reported in AML (31). These fusions typically place the EVI1 gene under the control of the fusion partner’s promoter, leading to its ectopic expression, and create a novel protein with combined or altered functions (Table 2).

One of the most studied fusions is the t(3;21)(q26;q22), which generates the AML1-EVI1 (also known as RUNX1-EVI1) fusion gene. This fusion is recurrently found in therapy-related MDS/AML and CML in blast crisis. The resulting AML1-EVI1 chimeric protein contains the N-terminal portion of the transcription factor AML1 (including its DNA-binding Runt domain) fused to the majority of the EVI1 protein (33, 38). This fusion protein has a dual, devastating impact. It acts as a dominant-negative inhibitor of wild-type AML1, a master regulator of normal hematopoiesis, thereby blocking differentiation (39). Simultaneously, it brings the potent oncogenic functions of EVI1 into the cell. The AML1-EVI1 fusion protein has been shown to efficiently transform hematopoietic stem cells and cause embryonic hematopoietic abnormalities in mouse models, underscoring its potent leukemogenic capacity (33, 40). The fusion protein also aberrantly sequesters the AML1 binding partner PEBP2β in the nucleus, further disrupting normal hematopoietic regulation (41).

Other translocations also create pathogenic EVI1 fusions. The t(3;12)(q26;p13) results in an ETV6-EVI1 (also known as TEL-EVI1) fusion, which is associated with CML and other myeloproliferative disorders with poor prognosis (42). The ETV6 component provides a potent promoter driving EVI1 expression and a self-association domain that may enhance the oncogenic activity of the fusion protein (43). Similarly, the t(2;3)(p15-22;q26) and t(3;6) can lead to EVI1 overexpression or fusion events associated with poor-risk AML (37). A fusion between RPN1 and EVI1 has also been described in patients with inv(3) or t(3;3), contributing to the adverse prognosis (21).

Beyond the classic GATA2 enhancer hijacking, EVI1 can be activated by co-opting other powerful regulatory elements. A compelling example is the hijacking of a MYC super-enhancer by the EVI1 locus in a subset of AML cases. This event leads to the formation of a neotad (a newly formed topologically associating domain) that places EVI1 under the control of a potent enhancer normally reserved for the MYC oncogene, resulting in massive EVI1 overexpression and aggressive leukemia (36). In T-cell acute lymphoblastic leukemia (T-ALL), translocations involving the T-cell receptor beta (TCRβ) locus can bring its powerful enhancer to the vicinity of EVI1, driving its expression and causing undifferentiated leukemia (44). These examples illustrate a common theme in EVI1 deregulation: its promoter is relatively weak and requires the illicit appropriation of strong, lineage-specific enhancers to achieve oncogenic levels of expression. Furthermore, comprehensive genomic analyses using RNA sequencing are continually improving the detection of both canonical and cryptic EVI1 rearrangements and fusion events in AML, providing a more complete picture of its activation landscape (45).

The historical discovery of EVI1 as an oncogene came from studies of murine leukemia models, where retroviral integration into the host genome served as a powerful tool for identifying cancer-causing genes. Insertions of retroviral long terminal repeats (LTRs), which contain strong promoter and enhancer elements, in or near the Evi1 gene locus were found to be a common event in myeloid leukemias induced by these viruses (3). This insertional mutagenesis provided the first direct evidence that aberrant activation of EVI1 expression is a primary oncogenic event. Similar insertional activation of the EVI1 homolog Prdm16 has also been shown to induce leukemia (46). These preclinical models have been instrumental in establishing the cause-and-effect relationship between EVI1 overexpression and leukemogenesis and continue to be valuable tools for studying its function (47).

A cardinal feature of EVI1-driven leukemia is a profound block in myeloid differentiation. EVI1 exerts this effect by repressing the expression and/or function of key transcription factors that are essential for normal hematopoiesis. One of its most critical targets is RUNX1 (AML1), a master regulator of definitive hematopoiesis. EVI1 directly binds to and represses the RUNX1 promoter and also physically interacts with the RUNX1 protein to inhibit its transcriptional activity, effectively shutting down its lineage commitment programs (48). This antagonism is a central node in EVI1-mediated leukemogenesis. Similarly, EVI1 interferes with differentiation induced by hematopoietic cytokines like granulocyte colony-stimulating factor (G-CSF) (19).

EVI1 also skews hematopoietic output. Its forced expression in hematopoietic progenitors can induce abnormal megakaryocytic differentiation, consistent with the thrombocytosis often seen in patients with 3q abnormalities (49). This is partly mediated by its influence on the TPO/MPL signaling pathway, which promotes the growth of EVI1-positive, CD41-positive megakaryocytic cells (50). At the stem cell level, EVI1 helps maintain the self-renewal and undifferentiated state of hematopoietic stem cells (HSCs), in part by upregulating the adhesion G-protein coupled receptor GPR56, which is critical for HSC maintenance (51). It also regulates the expression of other key hematopoietic regulators, including GATA-1, GATA-2, and SCL/TAL1 (52), and the transcription factor PLZF (53), further cementing its role as a global disruptor of the normal hematopoietic hierarchy. By blocking differentiation while promoting self-renewal, EVI1 effectively traps hematopoietic cells in an immature, proliferative state, a crucial step toward malignant transformation (9).

In addition to disrupting transcriptional networks governing differentiation, EVI1 promotes cell survival and proliferation by rewiring intracellular signaling pathways.

A primary mechanism is the potent inhibition of the Transforming Growth Factor-beta (TGF-β) signaling pathway, a critical tumor-suppressive pathway that normally induces cell cycle arrest and apoptosis. EVI1 accomplishes this through multiple interactions. It physically binds to Smad3, a key signal transducer of the TGF-β pathway, preventing it from participating in active transcriptional complexes (54). Furthermore, EVI1 recruits the transcriptional co-repressor C-terminal Binding Protein (CtBP) to the promoters of TGF-β target genes, leading to their active repression (55). The post-translational phosphorylation of EVI1 has been shown to enhance its interaction with CtBP, thereby strengthening its repressive function (56). By dismantling this crucial tumor suppressor checkpoint, EVI1 provides cells with a powerful survival and proliferative advantage. Interestingly, the AML1-EVI1 fusion protein also effectively blocks TGF-β signaling, contributing to its oncogenic activity (17).

EVI1 also promotes cell survival by activating the PI3K/AKT/mTOR pathway. It achieves this by transcriptionally repressing the tumor suppressor gene PTEN. The loss of PTEN function leads to the accumulation of PIP3 and subsequent constitutive activation of AKT and its downstream effector mTOR, a central regulator of cell growth, proliferation, and survival (57). More recently, EVI1 was shown to drive mTORC1 activation through a novel axis involving the upregulation of the histone demethylase Kdm6b, which in turn activates the expression of Laptm4b, a lysosomal protein that promotes mTORC1 signaling (58). This sustained activation of pro-survival signaling pathways makes EVI1-expressing cells highly resistant to apoptosis.

As a DNA-binding protein, EVI1’s core function is to reprogram the cellular transcriptome and epigenome. It acts as a scaffold for large protein complexes that modify chromatin and regulate gene expression (Table 3). EVI1 interacts with histone methyltransferases such as SUV39H1 and G9a, recruiting them to target gene promoters to mediate transcriptional repression through the deposition of repressive histone marks (e.g., H3K9 methylation) (59). This function is critical for its ability to immortalize bone marrow cells.

Beyond histone modifications, EVI1 has been shown to be a master regulator of the leukemic epigenome by directing aberrant patterns of DNA methylation. EVI1-high leukemias exhibit a distinct DNA hypermethylation signature, suggesting that EVI1 either recruits DNA methyltransferases to specific loci or represses genes involved in demethylation, thereby sculpting a pathogenic epigenetic landscape (69). Its oncogenic activity is also linked to its ability to repress the expression of the retinoblastoma (Rb) tumor suppressor gene, further disabling cell cycle control (70). In a more recently discovered mechanism, EVI1 has been shown to drive leukemogenesis through the aberrant transcriptional activation of the proto-oncogene ERG, highlighting its capacity to function as both a repressor and an activator (71). Through these widespread effects on the transcriptome and epigenome, EVI1 establishes and maintains the malignant state.

Leukemogenesis is a multi-step process, and EVI1 rarely acts alone. Its potent transforming activity is often realized through collaboration with other cooperating genetic mutations. EVI1 overexpression is a critical cooperating event in leukemias driven by MLL rearrangements, such as MLL-AF9. In this context, EVI1 is essential for the maintenance and growth of the leukemic clone, and its expression is required for the full transforming potential of the MLL fusion protein (72, 73).

Similarly, EVI1 cooperates powerfully with mutations in AML1 (RUNX1). Mouse models have demonstrated that the combination of Aml1 mutation and Evi1 overexpression is sufficient to induce a disease that faithfully recapitulates human MDS/AML (74, 75). This synergy is also observed in the progression of familial platelet disorder (FPD) with propensity to AML, a hereditary syndrome caused by germline RUNX1 mutations, where secondary acquisition of EVI1 overexpression is a key step in leukemic transformation (76).

EVI1 also synergizes with other oncogenic drivers. It has been shown to cooperate with a truncated, pro-leukemogenic isoform of C/EBPβ (LIP) to induce AML (77). It collaborates with the transcription factor Sox4 to activate retroviral LTRs and promote leukemogenesis (78). In concert with Trib1, EVI1 can accelerate AML development in the context of Hoxa9/Meis1 overexpression (79). Furthermore, EVI1 overexpression frequently co-occurs with activating mutations in the RAS signaling pathway, particularly in inv(3)/t(3;3) leukemias, suggesting a strong synergy between these two oncogenic pathways (80). The combination of oncogenic Nras and Evi1 has been shown to be a potent driver of aggressive leukemia in mouse models (81). This extensive network of collaborations underscores EVI1’s role as a potent and versatile oncogenic hub.

Across numerous studies and diverse subtypes of hematological malignancies, high EVI1 expression is consistently and independently associated with a dismal clinical outcome. In AML, EVI1 overexpression is one of the most significant adverse prognostic markers, predicting primary refractory disease, high rates of relapse, and poor overall survival (11, 82). This holds true for both pediatric and adult AML (10). Even within specific genetic subgroups, such as AML with MLL rearrangements, co-expression of EVI1 delineates a subset of patients with a particularly poor prognosis (83). EVI1-rearranged AMLs are recognized as a distinct entity characterized by a unique mutational and transcriptional signature and an exceptionally aggressive clinical course (84).

In the context of CML, EVI1 expression is rarely detected in the chronic phase but becomes significantly upregulated during progression to the myeloid blast crisis, where it is implicated as a key driver of transformation (85). Moreover, in CML patients who have developed resistance to tyrosine kinase inhibitors like imatinib, high EVI1 expression is a predictor of poor survival (86). Its expression is also linked to adverse outcomes in MDS and has been found to be upregulated in chronic lymphocytic leukemia (CLL), where it contributes to cell survival by modulating microRNA networks (87). The consistent association of EVI1 with aggressive disease across different malignancies solidifies its status as a top-tier negative prognostic biomarker.

Given its critical role in driving leukemogenesis and its association with therapy resistance, EVI1 itself represents a high-priority therapeutic target (Table 4). Recent studies have proposed several strategies to therapeutically target EVI1, including degradation of the EVI1 protein via the ubiquitin-proteasome pathway, inhibition of its transcriptional co-factors such as CTBP1/2 and HDACs, and interference with upstream regulators that control EVI1 expression or stability.

Arsenic trioxide (ATO), a drug used in the treatment of acute promyelocytic leukemia (APL), has been shown to induce the degradation of the EVI1 oncoprotein, leading to cell cycle arrest and apoptosis in EVI1-positive leukemia cells (88). ATO can also induce autophagy-related cell death in these cells by downregulating EVI1 (89). These findings suggest a potential repurposing of ATO for EVI1-driven malignancies.

More recent strategies focus on targeting the epigenetic machinery that EVI1 relies upon. BET bromodomain inhibitors, which disrupt the function of transcriptional co-activators, have been shown to suppress EVI1 expression and induce apoptosis in EVI1-high AML cells (66). Similarly, histone deacetylase (HDAC) inhibitors have demonstrated efficacy by downregulating EVI1 expression through a mechanism involving the up-regulation of its repressor, PA2G4 (90). The dependency of EVI1-high cells on specific metabolic pathways, such as those regulated by PRMT5, also offers a potential therapeutic vulnerability (91).

Direct inhibition of EVI1 has historically been challenging due to its nature as a transcriptional regulator lacking enzymatic activity. Consequently, several indirect strategies have shown promise in suppressing EVI1-driven leukemogenesis. Such approaches collectively aim to dismantle the oncogenic transcriptional network sustained by EVI1 rather than targeting the protein itself.

A key target gene of EVI1 in HSCs is GATA-2, a transcription factor crucial for HSC proliferation. EVI1 directly binds to the GATA-2 promoter and acts as an enhancer, upregulating its expression. This EVI1-GATA-2 axis is fundamental for the maintenance and proliferation of HSCs. The reduction in GATA-2 expression in Evi1^-/- HSCs underscores the hierarchical regulation of the HSC pool by transcription factors (92).

Furthermore, EVI1 can repress the activity of RUNX1, a critical hematopoietic regulator. This repression can impair RUNX1’s ability to bind DNA and regulate gene expression, potentially contributing to leukemogenesis by disrupting normal hematopoietic programs. The eighth zinc finger motif of EVI1 is involved in this interaction, and its expression alone can block granulocyte differentiation, leading to cell death. This suggests that inappropriate EVI1 expression can contribute to hematopoietic transformation by functionally impairing key hematopoietic regulators (48).

In the context of myeloid malignancies, EVI1 can inhibit the expression of the membrane-spanning-4-domains subfamily-A member-3 (MS4A3) gene. MS4A3 has been implicated in promoting differentiation in chronic myeloid leukemia (CML) by enhancing common β-chain cytokine receptor endocytosis. Low MS4A3 expression is characteristic of LSPC quiescence and transformation to blast phase CML, and EVI1’s suppression of MS4A3 contributes to the differentiation block observed in CML. Promoting MS4A3 re-expression or delivery of ectopic MS4A3 may be a therapeutic strategy to eliminate CML stem cells (93, 94).

Another approach involves targeting surrogate surface markers. It has been discovered that EVI1-high AML cells frequently express high levels of the surface protein CD52, which is not typically found on normal myeloid progenitors. This makes CD52 a potential target for antibody-based immunotherapy, such as with alemtuzumab, to specifically eliminate the EVI1-positive leukemic clone (95). Paradoxically, some studies have suggested that EVI1-positive AML cells may exhibit sensitivity to all-trans retinoic acid (ATRA), a differentiating agent, potentially through EVI1’s regulation by retinoic acid receptors (96, 97). These diverse approaches highlight the active search for effective therapies to combat these aggressive leukemias.