As noted above, the NSD histone methyltransferase family includes NSD1, NSD2, and NSD3, which regulate transcription through histone methylation and other mechanisms, such as by modulating the functions of transcription factors [16, 17, 28] as well as transcription-independent functions [18,19,20,21]. NSD1 methylates H3 lysine 36 (H3K36), contains a SET domain, zinc fingers, and nuclear localization signals, and its knockout causes embryonic lethality [29]. NSD2 has three isoforms, with the longest (NSD2-long) modifying H3K36 and playing a key role in transcription regulation [30]. NSD3 exists in multiple isoforms, where NSD3-whistle uniquely modifies H3K4 and H3K27, acting as a transcriptional repressor [31]. Mutations in NSD1 and NSD2 are linked to developmental syndromes and cancer, highlighting their critical biological functions [32, 33]. In addition, multiple chromosomal translocations have been identified in the context of NSD1, 2 and 3 with functional roles in cancer [22, 23, 34,35,36]. For a comprehensive overview of the cell-autonomous functions of NSD proteins in cancer, we refer readers to the reviews by Topchu et al. [37] and Bennett et al. [38]. Below, we discuss various roles that NSD proteins play in anti-tumor immunity.

Several previous studies have shown an important role for NSD1 in anti-tumor immunity [33, 39,40,41] (Fig. 1). A recent study uncovered a surprising mechanism of tumor immune evasion in head and neck squamous cell carcinomas (HNSCCs) involving the histone methyltransferase NSD1 [39]. NSD1 mutations induced DNA hypomethylation and retrotransposon de-repression. Both of these changes are typically associated with enhanced interferon responses and immune activation. However, NSD1-deficient HNSCCs paradoxically displayed an immune-cold phenotype [39]. Using both syngeneic and genetically engineered mouse models of HNSCC, the study demonstrated that NSD1 loss leads to immune exclusion and impaired interferon signaling, specifically through the silencing of key innate immune genes such as interferon lambda receptor 1 (IFNLR1). IFNLR1, also known as IL28RA, is a key component of the type III interferon (IFN-λ) receptor complex [42], and plays a critical role in innate immune responses [42].

NSD1 in anti-tumor immunity. (A) Loss of NSD1 in HNSCC induces epigenetic reprogramming via reduced H3K36me2 and increased EZH2-mediated H3K27me3, leading to silencing of innate immune genes such as IFNLR1. Despite DNA hypomethylation and retrotransposon de-repression, NSD1-mutant tumors display immune exclusion and suppressed interferon signaling. EZH2 inhibition restores immune infiltration and tumor control, revealing a druggable chromatin-based immune evasion mechanism. (B) NSD1 inactivation in HNSCC induces immune exclusion via epigenetic silencing of T-cell–attracting chemokines (CXCL9/10) through reduced H3K36me2 and increased H3K27me3. KDM2A inhibition restores chemokine expression, promotes T-cell infiltration, and suppresses tumor growth in an immune-dependent manner

Mechanistically, NSD1 loss disrupts the chromatin landscape by reducing H3K36me2 levels and enabling compensatory increases in H3K27me3, mediated by enhancer of zeste homolog 2 (EZH2), a histone methyltransferase of Polycomb Repressive Complex 2 (PRC2), which is known to cause transcriptional gene silencing. This epigenetic antagonism effectively shuts down the viral mimicry response and facilitates immune escape. Notably, treatment with an EZH2 inhibitor restores immune cell infiltration and inhibits tumor growth in NSD1-mutant models, highlighting a druggable chromatin crosstalk with potential therapeutic relevance [39]. These findings reshape our understanding of how chromatin modifiers influence tumor-immune dynamics and suggest targeting EZH2 as a viable strategy to sensitize NSD1-mutant HNSCCs to immunotherapy. These results are also consistent with several reports that have implicated increased EZH2 expression and activity in suppressing anti-tumor immunity in a variety of mouse models and cancer types, further highlighting its importance as a target for enhancing immunotherapy responses [43,44,45,46].

Another study classified HNSCC into three immune-based subtypes that included Immunity-High (H), Immunity-Medium (M), and Immunity-Low (L). Based on immune cell infiltration signatures this study revealed stark differences in tumor immunogenicity and response potential to immune checkpoint inhibitors (ICIs). The Immunity-H subtype exhibited high program cell death ligand-1 (PD-L1) expression, robust immune infiltration, low tumor heterogeneity, and favorable prognosis, making it more likely to benefit from immunotherapy. Conversely, Immunity-L tumors showed immune-cold features and poor clinical outcomes. Crucially, the authors found that mutations in chromatin regulators like NSD1 were enriched in the Immunity-H group and positively correlated with enhanced immune signatures, suggesting that alterations in NSD-family genes may promote anti-tumor immune activity. This is in contrast to the study described above, in which NSD1-mutations were associated with immune-cold tumors [39]. However, unlike the above study, this study was correlative and lacked functional validation. Nonetheless, another study supported the observation that NSD1-mutations creates an immune-cold environment. This study found that NSD1 mutations created an immune-cold phenotype in HNSCC and lung squamous cell carcinoma (LUSC). This immune-cold environment was characterized by reduced CD8+ T-cell and macrophage infiltration, lower programmed cell death protein 1 (PD-1)/PD-L1 expression, and higher ICI resistance [33].

Consistent with an immunosuppressive effect of NSD1, another study explored the epigenetic mechanisms by which NSD1 inactivation drives immune exclusion in HNSCC [41]. The authors found that loss of NSD1 resulted in reduced H3K36me2 and increased H3K27me3, which is a repressive histone mark, particularly on the promoters of key T-cell–attracting chemokines such as C-X-C Motif Chemokine Ligand 9 (CXCL9) and C-X-C Motif Chemokine Ligand 10 (CXCL10). As a result, NSD1-deficient tumors exhibited reduced expression of these chemokines, impaired T-cell infiltration, and resistance to PD-1 checkpoint blockade. This epigenetic silencing of immune effector genes contributes to the immune-cold tumor microenvironment often observed in NSD1-mutant HNSCC [41].

The study further identified lysine demethylase 2 A (KDM2A), a lysine demethylase that targets H3K36me2, as a druggable target to counteract the effects of NSD1 loss. Pharmacological or genetic inhibition of KDM2A restored H3K36me2 levels, reduced H3K27me3 at chemokine loci, and reinstated the expression of CXCL9 and CXCL10, leading to increased T-cell infiltration and suppressed tumor growth in immunocompetent mouse models. Notably, these effects were absent in immunodeficient mice, underscoring the immune-dependent mechanism of tumor control. The significance of this work lies in its demonstration that NSD1 inactivation reshapes the epigenetic landscape to evade immune surveillance, and that targeting KDM2A may represent a rational immunotherapeutic strategy to convert immune-cold tumors into immune-responsive ones. These findings position KDM2A inhibition as a novel epigenetic approach to enhance immunotherapy efficacy in NSD1-deficient cancers [41].

Collectively, the majority of studies on NSD1 suggest that NSD1 mutations promote an immune-cold tumor microenvironment, which may be reversed by targeting the altered chromatin landscape in these mutant cancers. While most functional investigations have focused on HNSCC, extending this analysis to other cancer types, particularly those harboring NSD1 alterations could reveal whether this immunosuppressive phenotype is conserved beyond HNSCC. Some evidence of this possibility comes from the study that is also described above in the context of LUSC [33].