Emerging role of microRNAs as modulators of the tumor immune microenvironment Authors: Rachele Frigerio, Arianna Ortolano, Gennaro Ciliberto, Rita Mancini, Luigi Fattore Published: 2026 ABSTRACT AbstractThe advent of immune checkpoint inhibitors (ICIs) has transformed the treatment landscape of several malignancies, but durable clinical benefits are only observed in a subset of patients. Primary and acquired resistance to immunotherapy, largely driven by an immunosuppressive tumor immune microenvironment (TIME), remains among the most critical hurdles to overcome. Amid the molecular regulators orchestrating immune evasion, tumor-derived microRNAs (miRNAs) have emerged as pivotal players. These small non-coding RNAs reshape the TIME through diverse mechanisms, including impairment of antigen presentation, modulation of immune checkpoint pathways, suppression of cytotoxic T and NK cell activity, induction of regulatory immune populations, and reprogramming of stromal and metabolic networks. This review explores the multifaceted contribution of miRNAs to ICI resistance and discusses innovative therapeutic strategies aimed at targeting dysregulated miRNAs, either through inhibition of oncomiRs or restoration of tumor-suppressors, with the goal of reprogramming the TIME toward an immunocompetent state. The clinical implementation of combination regimens of miRNA therapeutics with ICIs remains limited, highlighting the need to test these combinations in advanced preclinical models that faithfully recapitulate the human TIME. Such models could provide an optimal platform to integrate miRNA therapeutics into immuno-oncology strategies, ultimately expanding the benefits of ICIs to a broader patient population. Introduction Tumor growth, progression, and response to therapy are shaped by the dynamic, complex, and highly interconnected composition of the tumor immune microenvironment (TIME). The intricate interplay of diverse cellular and molecular mechanisms within the TIME plays a central role in the development of therapeutic resistance.1 2 Beyond malignant cells, the TIME consists of a variety of cellular components—including stromal, endothelial, and immune cells—as well as non-cellular elements such as the extracellular matrix, cytokines, growth factors, and metabolites, all of which can either support or suppress tumor progression.3 The TIME has emerged as a critical therapeutic target, particularly following the success of immune checkpoint inhibitors (ICIs), which have revolutionized cancer treatment by providing durable responses across multiple tumor types. These agents now constitute first-line therapies for several solid and hematologic malignancies. ICIs function by blocking interactions between inhibitory receptors on T cells and their ligands, thereby unleashing a robust immune activation and potent anti-tumor responses.35 To date, the US Food and Drug Administration has approved ICIs targeting three key immune checkpoints. The first ICIs approved for cancer therapy targeted cytotoxic T lymphocyte-associated protein 4 (CTLA-4) and the programmed cell death protein 1 (PD-1)/PD-L1 axis, were initially approved for a limited number of tumor types, most notably metastatic melanoma. Over time, the clinical use of these agents has expanded considerably, with multiple approved antibodies working against these targets in a wide range of malignancies.6 More recently, the immunotherapeutic landscape has further evolved with the introduction of a third class of ICIs targeting Lymphocyte Activation Gene 3 (LAG-3). Relatlimab, approved in 2022 in combination with nivolumab for metastatic melanoma, represents the first clinically available inhibitor of this immune checkpoint.6 All these therapies aim to reinvigorate T cell-mediated immunity,7 highlighting the central role of these cells, which are directly responsible for tumor cell killing.8 However, not all patients benefit from ICI therapy. Several patients exhibit primary resistance, while others develop acquired resistance following an initial response. These failures often arise due to the development of an immune-suppressive TIME. Consequently, only a subset of patients benefit from ICI treatment, and response rates remain suboptimal.1 5 Increasing evidence indicates that resistance to immunotherapeutic agents is not only driven by genomic or mutational alterations within the tumor, but also by non-genetic mechanisms that dynamically shape tumor–immune interactions. Among these, post-transcriptional regulatory processes are critical contributors of immune escape. In this context, microRNAs (miRNAs) are emerging as key regulators of antitumor immunity due to their pleiotropic nature enabling them to act on multiple targets and to co-ordinately modulate several immune-evasive pathways, including antigen presentation, immune checkpoint signaling, and metabolic adaptation,9 as it will be treated in detail in the section “MicroRNAs as modulators of resistance to immunotherapy” . This broad regulatory capacity highlights the importance of studying miRNA-mediated programs not merely as regulators of individual targets, but as modulators of interconnected immune networks. This systems-level mode of regulation also increases the therapeutic interest in miRNA-based strategies, as they may enable in theory the simultaneous modulation of multiple immune and tumor-associated pathways differently from the result of the selective inhibition of individual protein targets by specific inhibitors, whose effects may be circumvented by alternative pathways. This concept is particularly relevant in the context of the TIME, where the variability in ICI response is shaped by highly dynamic interactions between tumor, stromal, and immune components. The composition and functional state of the TIME are key determinants of both spontaneous anti-tumor immune activity and therapeutic efficacy. Immune suppression within the TIME, driven by various cellular and molecular determinants, including miRNA-mediated post-transcriptional programs that act as upstream regulators of immune networks, significantly impacts the outcome of immunotherapies. Therefore, reprogramming these components may help overcome TIME-induced resistance and enhance therapeutic responses.3 In the section “The time as a driver of immunotherapy resistance”, we first provide an overview of the TIME and the major mechanisms driving resistance to immunotherapy. In the second part of this review, we will examine the multifaceted role of tumor-derived microRNAs in modulating the TIME, and discuss their growing contribution to immune escape, resistance to ICIs, and their emerging potential as therapeutic targets. The time as a driver of immunotherapy resistance Every component of the TIME influences the development of an effective immune response in a positive or negative manner. In this section, we will provide an overview of the major mechanisms that contribute to immune evasion and resistance to immunotherapy, highlighting the key cellular players within the TIME that drive these processes. Malignant cells have developed an arsenal of mechanisms to evade immune surveillance. These include avoiding detection by immune effector cells, preventing the activation of immune cells even after successful recognition, and resisting cytotoxicity despite intact immune activation. A critical prerequisite for T cell-mediated cytotoxicity is the recognition of tumor-specific antigens. They are processed and presented by professional antigen-presenting cells (APCs), such as dendritic cells (DCs), macrophages, and B cells, which load tumor-derived peptides onto major histocompatibility complex class I (MHC-I) molecules. Peptide–MHC-I complexes, in combination with co-stimulatory molecules such as CD28, prime CD8+ T cells, which subsequently target and kill tumor cells expressing the same epitopes via their own MHC-I molecules. To circumvent this process, many tumors downregulate MHC-I expression.1012 In addition to altering antigen presentation on tumor cells, cancers may also disrupt antigen presentation function directly. This can occur through inhibiting APC recruitment, differentiation, or maturation, ultimately compromising their ability to initiate effective antitumor immunity. A key factor underlying this impaired interaction is the defective chemo-attraction between tumor cells and APCs.8 During immunogenic cell death, malignant cells release ATP and annexin A1 (ANXA1), which act as chemo-attractants to recruit APCs to the tumor site. However, many tumors develop mechanisms to suppress this recruitment. One such process involves the overexpression of CD39 on cancer cells, an ectonucleotidase that sequentially degrades extracellular ATP into ADP, AMP, and ultimately generates adenosine—a potent immunosuppressive metabolite. This enzymatic activity abolishes the ATP-mediated chemo-attraction of APCs, thereby impairing the initiation of adaptive immune responses.13 14 Since the efficacy of immune checkpoint blockade (ICB) relies on the presence of tumor-infiltrating lymphocytes, immunogenic cell death is also associated with the release of chemokines such as CXCL9 and CXCL10, which modulate the trafficking of immune effector cells to the TIME. High expression of these chemokines has been associated with improved patient survival and increased intratumoral T cell infiltration. However, tumors can evade immune surveillance by downregulating these chemokines, thereby impairing T cell recruitment and limiting the therapeutic efficacy of immunotherapy.15 In parallel, tumor and stromal cells can actively shape an immunosuppressive microenvironment through the secretion of cytokines such as CCL2. This chemokine recruits immunosuppressive populations, including Myeloid-Derived Suppressor Cells (MDSCs) and monocytes, further contributing to immune evasion and resistance to checkpoint blockades.8 In addition, another mechanism that malignant cells use to avoid recognition by T cells is physically excluding them from the core of the tumor through the establishment of structural barriers within the TIME. Central players in this immune-excluded phenotype are cancer-associated fibroblasts (CAFs) that release the transforming growth factor beta 1 (TGF-β1), which promotes the retention of CD8+ cytotoxic T lymphocytes within collagen-rich, CAF-dense peritumoral stromal regions. Consequently, CD8+ cytotoxic T lymphocytes are unable to penetrate the tumor core and to effectively exert their cytotoxic functions.16 When malignant cells are recognized by the immune system, they can still evade immune destruction by suppressing effector cell activity and promoting immunosuppressive cell populations. This is driven by alterations such as changes in immunomodulatory ligand expression, impaired pro-inflammatory cytokine signaling, and dysregulated metabolism.17 Within the TIME, malignant and immune cells compete for limited metabolic resources, as both rely on similar pathways to sustain their high bioenergetic and biosynthetic demands. This metabolic competition for key nutrients such as glucose, glutamine, methionine, tryptophan, and arginine creates a nutrient-depleted environment that impairs immune cell effector functions and weakens antitumor immunity. By depriving immune cells of these substrates, tumor cells create a metabolically hostile environment that promotes immune evasion and resistance to ICB.3 Once primed, activated, and recruited to the tumor, T cell function relies on a precise balance of stimulatory and inhibitory signals mediated by immune checkpoints—most notably PD-1/PD-L1, CTLA-4, LAG-3, and TIM-3—to execute their cytotoxic function. Tumor cells tilt this balance to favor their immune escape by upregulating checkpoint ligands, which deliver suppressive signals that blunt T cell activity and foster both primary and acquired resistance. Although therapies targeting a single checkpoint can be effective, tumors often activate compensatory alternative inhibitory pathways. Consequently, co-blocking multiple checkpoints has emerged as a rational strategy to counteract this adaptive upregulation and restore antitumor immunity.8 Moreover, cancer cells can evade immune-mediated cytotoxicity through various cytoprotective mechanisms that impair the formation or function of the immunological synapse—the interface where cytotoxic molecules such as perforin (PRF1) and granzyme B (GZMB) are released. These mechanisms include both mechanical and biological alterations that reduce cancer cell sensitivity to immune effector molecules. This reduced sensitivity to cytolysis by immune effector cells may result from impaired PRF1-mediated pore formation18 19 or from the triggering of pore repair,20 ultimately limiting the entry of GZMB into the cytosol, which compromises immune cell killing. A key player in orchestrating antitumor immunity and determinant of response to ICB is interferon-gamma (IFN-γ), which is predominantly secreted by activated CD8+ T cells, Th1 CD4+ T cells, and natural killer cells in response to IL-12 and IL-18 signaling. By engaging the JAK/STAT pathway, IFN-γ drives upregulation of MHC class I and II molecules, enhances antigen processing and presentation, promotes CD8+ T cell priming and activation, polarizes macrophages toward an M1 phenotype, and directly inhibits tumor cell growth and survival. From a critical perspective, genetic or functional defects in key IFN-γ pathway components abrogate these downstream effects and instead denote a well-established mechanism of primary as well as acquired resistance to ICIs.1 2 8 Even though immunotherapies primarily aim to boost T cell-mediated antitumor responses, their efficacy is often weakened by immunosuppressive cells in the TIME, such as regulatory T cells, MDSCs, and tumor-associated macrophages (TAMs). Poor clinical outcomes are typically associated with reduced effector T cell infiltration and an increase in these suppressive populations.1 21 Moreover, a high abundance of stromal components like CAFs correlates with poor prognosis. In particular, high TGF-β activity within the stroma has been linked to resistance to PD-L1 blockade, and its inhibition has shown potential in restoring antitumor immunity. These findings underscore the need to target both immune and stromal barriers to improve ICI efficacy.3 In addition to these barriers, the microbiota has recently emerged as another critical component involved in shaping the TIME. The intratumoral microbiome can activate inflammatory signaling pathways and sustain chronic inflammation, which progressively evolves into an immunosuppressive milieu that favors tumor progression by promoting immune cell dysfunction. Emerging studies highlight a close connection between the microbiota and T cell activity, largely mediated by microbiota-derived metabolites that modulate T cell metabolism and effector functions. These metabolites, such as short-chain fatty acids, enhance CD8+ T cell IFN-γ production, granzyme B expression, and memory differentiation, thereby strengthening cytotoxic responses. At the same time, intratumoral microorganisms can directly interact with tumor and immune cells, altering antigen presentation, cytokine secretion, and immune cell trafficking. Certain bacterial species suppress T cell–mediated cytotoxicity through activation of inhibitory receptors such as TIGIT and CEACAM1, whereas others promote the recruitment of IFN-γ-producing effector lymphocytes and potentiate the efficacy of ICB. Collectively, the microbiota acts as a dynamic regulator of antitumor immunity. Therefore, the microbiota and its metabolites are increasingly recognized as promising targets to enhance immunotherapy outcomes. For a more detailed overview relating to this subject, readers are welcome to refer to recent dedicated reviews.22 23 As briefly introduced above, tumor-derived miRNAs have emerged as key upstream regulators of the immunosuppressive TIME and important determinants of immunotherapy resistance. The section “MicroRNAs as modulators of resistance to immunotherapy” will examine the molecular mechanisms through which these regulatory molecules actively promote immune escape by impairing effector immune cell function, disrupting antigen presentation, and ultimately reducing tumor immunogenicity in more detail. MicroRNAs as modulators of resistance to immunotherapy The discovery of this class of small non-coding RNAs, which plays a fundamental role in gene regulation and homeostasis, and which is involved in disease development, in particular cancer, has recently been fully recognized by receiving the 2024 Nobel Prize award in Physiology and Medicine. The molecular mechanisms underlying miRNA biogenesis and their mechanism of action are widely documented in numerous exceptional recent reviews; therefore, they will not be addressed here.24 25 Depending on their targets, miRNAs may act either as oncomiRs, which promote uncontrolled gene proliferation, inhibit apoptosis, or induce cell migration and metastasis, or as tumor suppressors, which act by inhibiting oncogenic pathways. In some cases, a specific miRNA has been shown to behave either as an oncomiR or as an oncosuppressor depending on the cellular or tissue context where its deregulation has been observed.26 Furthermore, miRNAs are capable of modulating tumor immune surveillance and immune evasion. Certain miRNAs promote immune escape by reducing tumor cell immunogenicity and dampening anti-tumor immune responses, while others enhance immune-mediated tumor clearance.27 Both upregulation and downregulation of specific miRNAs can significantly influence tumor transformation and immune cell activity by altering the TIME, ultimately promoting malignancy.28 By dysregulating miRNA production, tumor cells may actively promote their own survival, evading apoptosis and enabling uncontrolled growth.29 MiRNA dysregulation in cancer cells is also spread to surrounding cells. Cancer-derived extracellular vesicles (EVs) enriched with oncomiRs and deficient in tumor-suppressive miRNAs facilitate the transfer of this altered miRNAome across the TIME.30 The following section will examine the role of miRNAs in tumor immune evasion, focusing on complex intracellular and extracellular pathways that enable malignant cells to evade immune surveillance. Dysregulated miRNAs contribute to immune escape by modulating antigen presentation, regulating immune checkpoint pathways, altering immune and stromal cell activity, as well as reshaping metabolic homeostasis, as shown in figure 1. MicroRNA-mediated regulation of the tumor immune microenvironment (TIME). A schematic overview of the multiple mechanisms through which miRNAs shape the TIME. miRNAs contribute to (i) the induction of immunosuppressive cell populations; (ii) CAFs activation and stromal remodeling; (iii) immune checkpoint regulation; (iv) suppression of effector immune cells; (v) impaired antigen presentation and DC function; and (vi) metabolic reprogramming. Through these interconnected pathways, miRNAs contribute to immune evasion and resistance to immunotherapy. CAFs, cancer-associated fibroblasts; CTLA-4, cytotoxic T-lymphocyte-associated protein 4; CXCL9, C-X-C motif chemokine ligand 9; DC, dendritic cell; GLUT1, glucose transporter type 1; LAG-3, lymphocyte activation gene 3; M1/2, macrophages type 1/2; MDSC, myeloid-derived suppressor cell; MHC-I/II, major histocompatibility complex class I/II; NK, natural killer; PD-1, programmed cell death protein 1; PD-L1, programmed death-ligand 1; SLC6A8, solute carrier family 6 member 8; TCR, T-cell receptor. Impaired antigen presentation and APC function Diverse tumor-derived miRNAs converge on the impairment of antigen presentation machinery, suggesting a recurrent post-transcriptional strategy by which tumors reduce immune recognition. A number of immune-regulatory miRNAs possess the capacity to target components of the major MHC-I.273134 For example, in esophageal adenocarcinoma (EAC) cell lines and tumor specimens, elevated expression levels of miR-125a-5p and miR-148a-3p have been observed. These miRNAs downregulate TAP2 and MHC-I—critical elements for effective antigen presentation. Consequently, EAC cells attenuate MHC-I-mediated antigen display, effectively excluding tumor-derived antigens from the cell surface and evading cytotoxic T lymphocyte-mediated immune surveillance.27 34 Another mechanism through which cancer cells evade cytolytic T cell recognition is mediated by miR-155. Upregulation of miR-155 in melanoma cells is associated with downregulation of microphthalmia-associated transcription factor-M (MITF-M), an isoform of MITF. This pivotal transcription factor regulates the expression of a variety of Tumor Associated Antigens genes, such as tyrosinase (TYR), Melan-A/Mart-1 (MLANA), or pMel17/gp100 (SILV), that are recognized by cytolytic T lymphocytes. Thus, the reduced expression of MITF-M due to the upregulation of miR-155 contributes to immune escape of melanoma cells.35 Beyond the tumor-intrinsic modulation of antigen presentation pathways, cancer cells also exploit miRNA-mediated mechanisms to interfere with the function of professional APCs, such as DCs, thereby further compromising the initiation of effective anti-tumor immunity.3638 For instance, Fabbri et al demonstrated that miR-21 and miR-29a, secreted by lung cancer cells via exosomes, can be internalized by DCs, where they directly bind to endosomal TLR7/8 through GU-rich motifs in their sequence. Acting as paracrine agonists, these miRNAs trigger TLR-mediated NF-κB activation, which induces the production of TNF-α and IL-6. The resulting inflammatory milieu promotes tumor growth and metastasis.37 Moreover, miR-212-3p, delivered by pancreatic cancer EVs, downregulates RFXAP, a transcription factor necessary for MHC class II expression.38 Immune checkpoint regulation MiRNAs are also implicated in the regulation of immune checkpoint networks, where both receptor and ligand expression are finely tuned across tumor and immune compartments, shaping a dynamic axis of immune inhibition. For instance, miR-15a-5p (miR-15a) and miR-15b-5p (miR-15b) function as tumor suppressors by directly targeting PD-L1 mRNA. This downregulation of PD-L1 facilitates an enhanced antitumor immune response, particularly in neuroblastoma.27 39 Similarly, in non-small cell lung cancer (NSCLC), miR-34 binds to the 3′ untranslated region of PD-L1, leading to a suppression of its expression.40 Other studies have found inverse correlations between miRNAs and immune checkpoint molecules. For example, miR-20b-5p is inversely correlated with CTLA-4 in renal cell carcinoma, while miR-424-3p targets both the CTLA-4/CD80 and PD-1/PD-L1 axes in prostate and ovarian cancers.31 41 MiRNAs can also regulate additional immune checkpoints such as LAG-342 and TIM-3.43 44 Case in point, miR-379-5p plays a key role in modulating CD8+ T cell activity within the TIME, enhancing their cytotoxic function by directly targeting and downregulating the immune checkpoint receptors TIM-3 and TIGIT. Notably, its expression is reduced in exhausted T cells, thus contributing to immune dysfunction. Operating as an epigenetic tumor suppressor, miR-379-5p represents a promising target to restore T cell responses and enhance the efficacy of cancer immunotherapy.44 MiRNAs have been shown to modulate PD-L1 expression also in macrophages.4548 For instance, in hepatocellular carcinoma, exosomes carrying miR-23a-3p downregulate PTEN in macrophages, leading to AKT phosphorylation and increased PD-L1 expression, which suppresses anti-tumor T cell responses.48 Suppression of effector immune cells MiRNAs contribute to drug resistance by disrupting the activity of immune effector cells. They converge on pathways that drive the functional exhaustion of cytotoxic lymphocytes and the survival and cytolytic capacity of effector populations, thereby weakening antitumor immunity. Conversely, specific miRNAs may enhance immune recognition and contribute to more effective tumor control, highlighting their dual role in shaping the tumor–immune interface and their potential for therapeutic reprogramming. Wang et al demonstrated that high levels of miR-155 in human breast tumors correlate with favorable antitumor immune infiltration and improved patient prognosis. In murine models, miR-155 overexpression increased T cell infiltration, delayed tumor growth, and enhanced tumor sensitivity to ICB therapy. From a mechanistic perspective, this was mediated through the upregulation of CXCL9/10/11 production via SOCS1 inhibition, resulting in increased p-STAT1/p-STAT3 ratios and promotion of T cell infiltration. These findings suggest that the therapeutic elevation of miR-155 in breast tumors may improve the efficacy of ICB therapy via remodeling the antitumor immune landscape.49 Another mechanism that influences immune responses involves the microRNA-183/96/182 cluster (m96cl). This cluster enhances CD8+ T cell proliferation and functions through a paracrine stimulation of IL-2 signaling. However, this mechanism is often downregulated in several cancers, including NSCLC. m96cl regulates IL-2 expression by targeting the transcriptional repressors Foxf2 and Zeb1. Restoring its expression can trigger a robust antitumor immune response by promoting CD8+ T cell expansion and activation via paracrine mechanisms.50 While therapeutic strategies for miR-155 and the miR-183/96/182 cluster would aim to restore or enhance their expression, other miRNAs should instead be therapeutically suppressed, as they exert opposite, pro-tumorigenic effects that directly promote immune evasion by impairing the activity and survival of effector immune cells.5153 As evidenced by Vignard et al, a panel of miRNAs such as miR-3187-3p, miR-498, and miR-181a/b-enriched in melanoma-derived exosomes that contribute to immune escape. These miRNAs impair T cell receptor (TCR) signaling and reduce cytokine secretion (eg, TNF-α), both of which are crucial for effective cytotoxic T cell activity.53 MiRNAs also play a pivotal role in modulating NK cell-mediated immune responses through diverse mechanisms.5456 Notably, bladder cancer-derived exosomes enriched in miR-221-5p and miR-186-5p have been shown to compromise NK cell intrinsic cytotoxic functions by destabilizing mRNAs encoding key cytotoxicity-related molecules, including DAP10, CD96, and perforin, consequently reducing their viability and killing capacity.56 Induction of immunosuppressive cell populations Tumor-derived miRNAs play a coordinated role in shaping a permissive TIME by promoting the expansion and functional polarization of immunosuppressive cell populations, including Tregs, TAMs, and MDSCs. Among these, regulatory T cells play a vital role in modulating immune responses and maintaining self-tolerance. Their importance in immune regulation was highlighted by awarding the 2025 Nobel Prize in Physiology or Medicine to Mary Brunkow, Fred Ramsdell, and Shimon Sakaguchi for the discovery of Tregs. Tumor cells can enhance Tregs recruitment in order to escape immune attack, partly through the secretion of tumor-specific miRNAs such as miR-214 via EVs. On uptake by peripheral CD4+T cells, miR-214 downregulates PTEN, promotes Treg expansion, and increases IL-10 production, thereby reinforcing immune suppression and supporting tumor progression.57 Treg expansion is also facilitated by cancer cell-derived miR-208b in colon cancer. This is transferred to T cells where it targets and downregulates programmed cell death factor 4 (PDCD4), further contributing to immune suppression.58 Another immunosuppressive cell population is represented by TAMs. These cells are often broadly categorized into M1-like and M2-like phenotypes, with M1 macrophages described as anti-tumorigenic and M2 macrophages as exhibiting immunoregulatory properties that may support tumor growth. However, this rigid classification is now seen as an oversimplification, since TAMs are highly plastic and can adopt a range of functional states in response to cues from the TIME. The plasticity of macrophages within the TIME has been widely studied, with compelling evidence indicating that both TAM-derived and tumor-derived miRNAs promote macrophage polarization, shifting them from pro-inflammatory to immunosuppressive.5965 A prominent example is provided by Sahraei et al, who showed that miR-21 in TAMs sustains an immunosuppressive phenotype. Consistently with this, its inhibition reprograms TAMs to produce IL-12 and CXCL10, boosting CD8+ T-cell activation, favoring tumor growth suppression. This identifies TAM-derived miR-21 as a potential target to enhance antitumor immunity and improve immunotherapy efficacy.63 Beyond direct cellular expression, exosome-mediated miRNA transfer also plays a critical role in this polarization. In glioblastoma, for instance, exosomal miR-374b-3p targets PTEN in macrophages, inducing M2 polarization and stimulating the secretion of TGF-β and MMP-9, thus promoting tumor angiogenesis.64 Exosomal miR-1246 from glioma cells has also been shown to induce M2 macrophage polarization by targeting TERF2IP, activating STAT3 signaling and inhibiting NF-κB pathway, thereby facilitating the malignant progression of glioma cells.65 Tumor-derived miRNAs also play a central role in the expansion and functional activation of MDSCs.6668 A compelling example of this mechanism is described by Huber et al68 in metastatic melanoma, where melanoma-derived EVs transfer a distinct set of miRNAs—including miR-146a, miR-155, miR-125b, miR-100, let-7e, miR-125a, miR-146b, and miR-99b—that reprogram CD14+ monocytes into MDSCs. These cells are characterized by HLA-DRA downregulation and increased production of IL-6 and CCL2, which promote immunosuppression and tumor growth.68 Metabolic reprogramming of the TIME Increasing evidence indicates that miRNAs orchestrate a complex metabolic reprogramming within the TIME, establishing a competitive landscape that favors tumor evasion. This is achieved through diverse mechanisms such as the promotion of tumor glycolysis and the impairment of immune cell metabolic fitness. In cancer cells, miRNAs contribute to the promotion of the Warburg effect—a well-known adaptive metabolic shift in cancer cells where they undergo enhanced glucose uptake and conversion to lactate, even in the presence of oxygen, rather than utilizing the mitochondria for energy production.69 This “aerobic glycolysis” supports rapid cancer cell growth and proliferation. The resulting lactate accumulation contributes to an acidic tumor milieu, which in turn promotes immune evasion, angiogenesis, invasion, and metastasis.27 Given their elevated proliferative demands, cancer cells require increased nutrient uptake. Emerging evidence indicates that miRNAs can directly or indirectly modulate the activity of metabolic enzymes involved in this reprogramming.70 For example, Xu et al demonstrated that in Oral Squamous Cell Carcinoma, the downregulation of miR-218 enhances cellular proliferation and glucose uptake through the upregulation of GLUT1 expression,71 thereby contributing to a TIME in which glucose becomes scarce for immune effector cells. Similarly, cancer-derived miRNAs can also directly impair the metabolic fitness of immune cells: in acute myeloid leukemia, leukemic cells release miR-19a-3p into the TIME via small EVs, which are subsequently internalized by CD8+T cells. This miRNA directly downregulates SLC6A8, a creatine transporter essential for ATP production, ultimately leading to impaired CD8+T cell metabolism and function.72 CAF activation and stromal remodeling Non-immune stromal cells within the TIME, CAFs, are also targeted by miRNAs. Emerging evidence indicates that miRNAs actively induce the differentiation of resident fibroblasts into pro-tumorigenic CAFs or enhance the pro-tumorigenic features of existing CAFs. In breast cancer, miR-222-3p has been shown to induce a CAF-like pro-tumorigenic phenotype in normal fibroblasts. Mechanistically, miR-222-3p targets the Lamin B receptor (LBR), leading to the induction of cellular senescence and the acquisition of a senescence-associated secretory phenotype (SASP). While senescence is traditionally considered a tumor-suppressive process, increasing evidence indicates that senescent fibroblasts can paradoxically promote tumor progression through the secretion of SASP factors.73 In line with this, miR-222-3p-mediated LBR downregulation in normal fibroblasts enhances the secretion of SASP components, including the classical senescence marker MMP3 and the pro-inflammatory cytokines IL-6 and IL-8. These factors contribute to fibroblast reprogramming toward a CAF-like state, thereby fostering tumor growth, invasion, and epithelial–mesenchymal transition.74 Similarly, tumor-derived EVs have been shown to reprogram fibroblasts into CAFs or enhance their activated state in other types of cancer.7577 For instance, melanoma cell-derived EVs induce conversion of normal fibroblasts into CAFs via miR-92b-3p mediated downregulation of PTEN.77 In summary, all these studies, summarized in table 1, underscore the multifaceted ways in which miRNAs regulate immune evasion. Depending on the context, they can either enhance immune-driven tumor suppression or promote immune evasion by impairing effector cell activity through either direct intracellular mechanisms and exosomal transfer. Understanding this dual role is crucial for designing miRNA-based therapeutic strategies to overcome resistance and improve cancer immunotherapy outcomes. Table 1miRNAs regulators of tumor immune evasion.miRNAsCancerMechanism/effectImmune cell types involvedReferencesAntigen presentation defectsmiR-125a-5p, miR-148a-3p, miR-21-3p, miR-200a, miR-155, miR-21*, miR-29a*, miR-203*, miR-212-3p*EAC, Melanoma, BC, Pancreatic cancerDownregulation of TAP1/2, MHC-I, MITF-M, TLR4/7/8 → Impaired antigen presentation and DC maturation, RFXAP/MHC-II suppressionDCs, CD4+ T helper cells 2732 38 Immune checkpoint regulationmiR-15a/b, miR-34, miR-23a-3p, miR-21-5p*, miR-200a*, miR-106b-5p*, miR-18a-5p*, miR-27a-3p, miR-20b-5p, miR-424-3p, miR-93-5p, miR-379-5pNB, NSCLC, HCC, CRC, BC, RCC, Prostate/ovarian cancersDirect binding to PD-L1, CTLA-4, TIM-3, TIGIT → upregulation or downregulation of checkpoint activityCD8+ T cellsTAMs 39 4143 Effector immune cell modulationmiR-155, miR-183/96/182 cluster, miR-20a-5p*, miR-3187-3p, miR-498, miR-181a/b*, miR-690*, miR-677*, miR-1957a*, miR-519a-3p, miR-92a, miR-221-5p*, miR-186-5p*BC, NSCLC, TNBC, Melanoma, Glioma, Bladder cancerImpaired TCR signaling, reduced cytokine secretion, apoptosis induction, reduced NK recognitionCD8+ T cells, NK cells,CD4+T cells 49 56 Induction of immunosuppressive cellsmiR-214*, miR-208b*, miR-21, miR-21-5p*, miR-103a*, miR-374b-3p*, miR-1246*, miR-146a/b*, miR-155*, miR-125a/b*, miR-100*, miR-99b*, let-7e*, miR-107*, miR-29a*, miR-92a*Colon cancer, HCC, HNSCC, lung cancer, glioblastoma, glioma, melanoma, gastric cancerPTEN/STAT3 axis activation, polarization toward immunosuppressive phenotypesTregs, TAMs, MDSCs 57 68 Metabolic remodelingmiR-218, miR-19a-3p*OSCC, acute myeloid leukemiaUpregulation of GLUT1 (enhanced glucose uptake of cancer cell), downregulation of SLC6A8 (impaired ATP production in T cells)CD8+ T cells 71 72 Stromal remodeling (CAF activation)miR-222, miR-92b-3p*, miR-21*, miR-27a*BC, Melanoma, HCC, gastric cancerInduction of CAF phenotype via PTEN/AKT, LBR, CSRP2 targetingCAFs 74 77 *extracellular vesicles-derived miRNAsBC, breast cancer; CAFs, cancer-associated fibroblasts; CRC, colorectal cancer; DC, dendritic cell; EAC, esophageal adenocarcinoma; HCC, hepatocellular carcinoma; HNSCC, head and neck squamous cell carcinoma; MDSCs, myeloid-derived suppressor cells; MHC-I, major histocompatibility complex class I; miRNAs, microRNAs; MITF-M, microphthalmia-associated transcription factor-M; NB, neuroblastoma; NK, natural killer; NSCLC, non-small cell lung cancer; OSCC, oral squamous cell carcinoma; RCC, renal cell carcinoma; TAMs, tumor-associated macrophages; TNBC, triple negative breast cancer. miRNAs regulators of tumor immune evasion. extracellular vesicles-derived miRNAs BC, breast cancer; CAFs, cancer-associated fibroblasts; CRC, colorectal cancer; DC, dendritic cell; EAC, esophageal adenocarcinoma; HCC, hepatocellular carcinoma; HNSCC, head and neck squamous cell carcinoma; MDSCs, myeloid-derived suppressor cells; MHC-I, major histocompatibility complex class I; miRNAs, microRNAs; MITF-M, microphthalmia-associated transcription factor-M; NB, neuroblastoma; NK, natural killer; NSCLC, non-small cell lung cancer; OSCC, oral squamous cell carcinoma; RCC, renal cell carcinoma; TAMs, tumor-associated macrophages; TNBC, triple negative breast cancer. These observations acquire particular relevance in light of the current immunotherapeutic landscape. Even though ICIs targeting CTLA-4, PD-1/PD-L1, and LAG-3 have been approved for a wide range of malignancies, resistance still remains a major clinical challenge. As discussed above, increasing evidence suggests that tumor-specific miRNA dysregulation represents an additional regulatory layer capable of shaping response to these agents. Distinct tumor types are characterized by defined oncomiRs and tumor-suppressive miRNAs that coexist with approved ICI regimens, indicating the presence of miRNA-driven immunosuppressive circuits that may limit therapeutic efficacy. In this scenario, combining ICB with strategies aimed at inhibiting oncogenic miRNAs or restoring tumor-suppressive miRNAs appears as a rational approach toward dismantling resistance mechanisms and enhancing antitumor immune responses. Figure 2 summarizes the ICIs that have currently been approved for clinical use across various major tumor types, alongside representative miRNAs involved in immune modulation, thereby contributing to ICI resistance in these malignancies. FDA-approved immune checkpoint inhibitors (ICIs) across major solid tumor types and representative microRNAs associated with resistance to immunotherapy. For each malignancy, the approved ICIs are categorized according to their molecular targets (anti-CTLA-4, anti-PD-1, anti-PD-L1, and anti-LAG-3). Representative oncomiRs (in red) and tumor-suppressive miRNAs (in green) reported in the literature as involved in modulating response and resistance to ICIs are shown for each tumor type. CTLA-4, cytotoxic T-lymphocyte-associated protein 4; FDA, Food and Drug Administration; LAG-3, lymphocyte activation gene 3; PD-1, programmed cell death protein 1; PD-L1, programmed death-ligand 1. A detailed discussion of how miRNA-targeting strategies can reshape the TIME by promoting the transition from an immunosuppressive to an immunocompetent and immunotherapy-responsive state will be provided in the section “Targeting miRNA dysregulation to restore an anti-tumoral TIME”. Targeting miRNA dysregulation to restore an anti-tumoral TIME Mechanisms of action and preclinical evidence of miRNA-based immunotherapeutic approaches The modulation of dysregulated miRNA expression is capable of reshaping the TIME into a more immunocompetent state. An example of this was first proposed by Kim et al and consists in the targeting of miR-21, one of the first oncogenic miRNAs ever to be identified. miR-21 is frequently overexpressed in various tumor types as well as being implicated not only in the direct promotion of cancer cell survival and expansion but also in the modulation of the TIME, particularly in promoting macrophage polarization toward a pro-tumorigenic M2-like phenotype. In their study, the authors developed a dual-targeted drug delivery system designed to deliver miR-21 inhibitors to specific tumor cells and macrophages overexpressing PD-L1. This system, termed Pep-21, comprises a PD-L1-binding peptide covalently linked to an anti-miR-21 oligonucleotide via click chemistry. Pep-21 was selectively internalized by both tumor cells and macrophages, effectively depleting endogenous miR-21. This led to a reduction in tumor cell migration and a reprogramming of immunosuppressive M2-type TAMs into pro-inflammatory M1-type macrophages, as well as an overall inhibition of tumor progression.78 In a subsequent study, the same group further elucidated the mechanisms by which miR-21 inhibition reverses immunosuppression within the TIME and stimulates antitumor immunity. The authors demonstrated that silencing miR-21 activates signaling pathways associated with immunogenic cell death, via the induction of endoplasmic reticulum stress and autophagy. Moreover, reduced miR-21 expression was shown to promote DC maturation, with the enhancement of CD8+T cell-mediated immune responses. More importantly, the combination of miR-21 inhibition with anti-PD-L1 therapy significantly improved therapeutic outcomes, resulting in a decrease in primary tumor growth and a reduction in lung metastasis in vivo.79 The oncogenic activity of miR-21 reported above is counterbalanced by other evidence showing that in certain conditions, the same miRNA may instead act as a tumor suppressor. For example, Guo et al investigated the role of miR-21-3p in promoting ferroptosis in melanoma cells in response to IFN-γ secreted by CD8+ T lymphocytes. Their findings demonstrated that the upregulation of miR-21-3p enhances IFN-γ-induced ferroptosis by amplifying lipid peroxidation. Furthermore, systemic delivery of miR-21-3p using gold nanoparticles significantly improved the therapeutic efficacy of anti-PD-1 antibody treatment in a preclinical mouse model, without inducing notable side effects.80 Another study identified miR-182 as a key regulator of tumor-induced M2 polarization and a promising target for therapeutic macrophage reprogramming.81 Mechanistically, it was shown that tumor cells induce miR-182 expression in macrophages via TGF-β signaling, which in turn suppresses TLR4 expression, thereby inhibiting NF-κB activation and promoting M2 polarization. Targeted delivery of antagomiR-182 via cationized mannan-modified EVs efficiently suppressed miR-182 expression, leading to macrophage reprogramming and tumor suppression in multiple murine models of BC. Notably, miR-199a-5p and miR-204-5p, two tumor-suppressive miRNAs found to be downregulated in drug-resistant melanoma cells, have shown a dual anti-tumoral effect. Lipid nanoparticles (LNPs) co-encapsulating these miRNAs not only directly inhibit melanoma cell growth in vitro and in vivo, but also exert an indirect effect on the tumor microenvironment by impairing the recruitment and reprogramming of pro-tumoral macrophages. This is achieved through the downregulation of several pro-angiogenic and pro-inflammatory factors released by melanoma cells. At the molecular level, these effects are mediated by the disruption of a newly identified miR-204-5p/miR-199b-5p–CCL5 axis, further emphasizing the relevance of miRNA-mediated regulation of macrophage dynamics in cancer progression.82 Another approach aiming to restore the expression of tumor-suppressive miRNAs and to enhance immune responses has been proposed by Taghikhani et al, who engineered tumor-derived exosomes (TEX) to overexpress miR-155, miR-142, and let-7i—miRNAs known to regulate DC maturation and function. These modified TEX demonstrated improved immunostimulatory capacity and antitumor activity, likely through their capacity to modulate the immunosuppressive TIME.83 Collectively, these studies highlight the therapeutic potential of targeting dysregulated miRNAs to modulate the TIME and enhance the efficacy of ICIs. Translational challenges and limitations of miRNA-based immunotherapeutic strategies Despite the encouraging preclinical evidence supporting miRNA-based strategies to reshape the TIME, their clinical translation remains hampered by several critical challenges. One of the first characteristics is the intrinsic pleiotropic nature of miRNAs. This aspect may represent a conceptual advantage, as individual miRNAs are able to concurrently regulate multiple signaling pathways and resistance mechanisms, thereby allowing the coordinated modulation of interconnected oncogenic and immunosuppressive networks. However, this same pleiotropy constitutes a major translational limitation, as systemic modulation of miRNA activity may result in unintended off-target effects across multiple cell types, including non-malignant immune populations within the TIME, thereby increasing the risk of unpredictable biological and immunological outcomes. For this reason, prior to targeting miRNAs with established pro-tumor roles in cancer cells, their functions in the tumor microenvironment must be carefully evaluated. A comprehensive testing of the effects of these miRNAs will help elucidate context-dependent miRNA functions and facilitate the identification of suitable targets for cancer therapy.84 An additional challenge for the clinical translation of miRNA-based therapeutics arises from an inherent instability and unfavorable pharmacokinetic properties of RNA-based molecules, as naked miRNAs are highly susceptible to nuclease degradation and exhibit limited cellular uptake and endosomal escape. To address these limitations, advanced delivery technologies have been developed. LNPs are among the most clinically advanced RNA delivery systems that provide protection from degradation and efficient intracellular delivery. However, their clinical application remains challenging due to systemic toxicity, limited tissue specificity, and the potential to trigger immune activation. Indeed, exogenous miRNAs can activate innate immune sensing pathways through pattern-recognition receptors, leading to systemic inflammation and immune-related adverse events. These liabilities were highlighted by the early termination of clinical trials involving miRNA mimics, most notably MRX34,85 due to severe immune-mediated toxicities. More recently, engineered EVs and exosomes have gained interest as alternative delivery vehicles due to their intrinsic biocompatibility, reduced immunogenicity, and natural ability to transfer RNA cargo between cells. Exosomes can also be modified to achieve cell type-specific targeting.86 Nonetheless, challenges related to scalable production, efficient cargo loading, and manufacturing standardization currently hinder their broad clinical translation.87 88 Furthermore, supra-physiological levels of miRNA mimics may saturate components of the endogenous RNA interference machinery, thereby perturbing endogenous miRNA regulatory networks and amplifying off-target transcriptomic effects. In particular, distinct miRNA-targeting strategies pose different translational challenges. In contrast to miRNA replacement approaches, oncogenic miRNAs can be inhibited by using antisense oligonucleotides (ASOs), which in some contexts can be administered without complex delivery vehicles, thereby potentially reducing carrier-associated toxicities. However, their limited membrane permeability and susceptibility to nuclease degradation require chemical modifications to improve stability, cellular uptake, and pharmacokinetic properties.89 In this context, locked nucleic acid (LNA)-modified ASOs have emerged as a particularly robust platform, showing enhanced stability, high target affinity, and a favorable safety profile, as also exemplified by their clinical success in a phase I trial.90 However, to date, LNA-based miRNA inhibitors have not yet been investigated within the context of ICIs resistance. All together, these considerations underscore that, while miRNA-based modulation of the TIME offers a unique opportunity to enhance the efficacy of immunotherapies, successful clinical translation will require refined delivery strategies, through target selection and careful dosing of therapeutics to preserve physiological miRNA networks in order to improve stability and limit toxicity, as summarized in table 2. Table 2Key translational challenges and potential solutions for miRNA-based therapeuticsChallengesPotential solutionsRNA instability and nuclease degradationAdvanced chemical modification (LNAs) and encapsulation in lipid nanoparticlesPoor pharmacokinetics and cellular uptakeOptimization of formulations to enhance cellular uptake and endosomal escapeSystemic toxicity and immunogenicity of delivery vehiclesOptimization of lipid composition, use of engineered extracellular vesicles, use of non-lipid-based systems (eg, LNAs)Saturation of endogenous RNA interference machineryCareful dose optimization to preserve physiological miRNA networksOff-target effectsDevelopment of cell-type-specific delivery approachesLNA, locked nucleic acid; miRNA, microRNA. Key translational challenges and potential solutions for miRNA-based therapeutics LNA, locked nucleic acid; miRNA, microRNA. In this context, the availability of robust preclinical models that accurately recapitulate the immunological complexity of human tumors would represent a critical intermediate step for addressing stability, safety, and efficacy concerns. To this end, there is an ever-increasing emphasis on the development of advanced, patient-derived in vitro systems that incorporate immune components of the TIME, offering a more physiologically relevant platform for evaluating novel immunotherapeutic interventions and for elucidating mechanisms of response and resistance. Preclinical models suitable for testing new immunomodulatory therapies Despite the encouraging initial findings, the approach of combining miRNA targeting agents with ICIs has not yet been translated into clinical trials. This gap is largely due to the need to provide more robust evidence of efficacy in highly predictive preclinical models. Conventional models, including in vitro 2D culture, 3D spheroids, and in vivo PDXs, poorly recapitulate the heterogeneous TIME of native tumors,91 and are often not predictive of clinical outcomes.92 In this regard, patient-derived organoids (PDO) and ex vivo tissue systems could represent powerful platforms to dissect mechanisms of resistance, to better validate miRNA targets and to optimize miRNA-based immunomodulatory strategies in a personalized setting. In fact, developing models that faithfully reproduce patient-specific TIME would improve patient selection, target identification, and the definition of resistance mechanisms for immuno-oncology therapeutics.93 Preserving the native TIME is critical, as it plays a central role in tumor progression, metastasis, and carcinogenesis. Among its components, the immune compartment, especially the T cell effector population, remains the most difficult to maintain in vitro. Tumor organoids, 3D culture systems which retain tissue architecture and cellular heterogeneity, provide a promising solution.94 To better preserve immune cells, particularly T cells, three main approaches have been developed to date, as described below and schematically illustrated in figure 3, which also outlines their potential for studying immunotherapy in combination with miRNA-targeted approaches, with reference to miRNAs previously discussed in this review. Schematic representation of three preclinical models for studying immunotherapy in combination with miRNA-targeted approaches. Examples include combinations of ICIs with: (i) an miR-155 agonist enhancing CXCL9 expression and T cell recruitment, (ii) an miR-125 antagonist restoring MHC-I expression and T cell cytotoxicity, and (iii) antagonists of miR-181a/b and miR-498 that restore TCR signaling and cytokine secretion (ie, TNF-α) for cytotoxic T cell activity. MHC-I, major histocompatibility complex class I; PBLs, peripheral blood lymphocytes; PD-1, programmed cell death protein 1; PD-L1, programmed death-ligand 1; PDOs, patient-derived organoids; PDTFs, patient-derived tumor fragments; TCR, T-cell receptor; TILs, tumor-infiltrating lymphocytes. The first model consists of PDOs, developed from single-cell suspensions of patient tumors in matrigel or collagen gel, that preserve both inter- and intra-tumoral heterogeneity and diverse immune cell populations, that enable testing of anti-PD-1 antibodies, tumor-infiltrating lymphocytes (TILs), and small molecules, as well as assessing TIL chemotaxis and cytotoxicity.95 Another approach is represented by reconstituting immune organoids, generated by co-culturing autologous tumor organoids with peripheral blood lymphocytes, that allows for the expansion of tumor-reactive T cells and the evaluation of their effector functions, providing a personalized approach for T-cell-based immunotherapy.96 97 Lastly, patient-derived tumor fragments (PDTFs) represent an organotypic culture system in which fresh tumor tissues from surgical resections are dissected into fragments that maintain tissue architecture and cellular complexity while ensuring sufficient diffusion of nutrients and reagents. One distinguishing advantage of this approach, in which, in our opinion, offers significant advantages over the previous ones, is its capacity to mitigate the confounding effects of tumor heterogeneity through including fragments from multiple tumor regions. These conditions permit ex vivo perturbation studies to assess early immune responses to checkpoint blockades and offer predictive insights for patient stratification and biomarker development.98 By more faithfully reproducing the tumor microenvironment, these models offer a unique platform to evaluate combined treatments, for example, pairing ICIs therapies with miRNA-targeted interventions. To date, no study has systematically exploited PDOs or PDTFs platforms as dedicated systems to investigate miRNA-based tumor immunomodulation. While these models are increasingly used for immunotherapy testing and drug response prediction, their integration with miRNA-based approaches remains largely unexplored. Nevertheless, the complementary features of PDOs and PDTFs make them conceptually well suited for distinct stages of miRNA-based immunomodulatory research. PDOs provide a scalable and experimentally flexible system that can support screening of miRNA mimics, antagomirs, or delivery formulations, making them well suited for comparative target validation and dose–response studies.96 97 Nonetheless, the dissociation and reaggregation steps required for PDOs establishment often result in partial loss of the native TIME, particularly affecting spatial immune organization and stromal–immune interactions, which are critical for studying miRNA-regulated processes implicated in T-cell exclusion, immune desertification, or fibroblast-mediated immunosuppression.91 99 In contrast, PDTFs preserve native tissue architecture and spatial tumor–immune interactions, offering a more faithful, although temporally limited, representation of patient-specific TIME.98 Moreover, multiregion sampling partially mitigates intratumoral heterogeneity, a key confounder in immunotherapy response assessment.93 The main limitation of this approach is the reduced scalability and shorter lifespan of ex vivo cultures, which limit their applicability for large-scale screening but make them highly informative for mechanistic and predictive studies. Most importantly, PDOs and PDTFs may also differ in their ability to recapitulate intratumoral delivery barriers relevant to miRNA-based therapeutics. Dense extracellular matrix and 3D tissue architecture impose physical constraints that limit therapeutic diffusion and cellular uptake in vivo.100 In conventional PDO cultures, exposure to therapeutic agents occurs through diffusion from the culture medium, within an artificial extracellular matrix environment that lacks the native stromal organization and mechanical properties of tumors. This experimental setup makes them suitable for comparative evaluation and optimization of delivery strategies under controlled conditions, but may underestimate diffusion limitations and spatial heterogeneity associated with matrix density and tissue architecture in vivo.91 101 Conversely, PDTFs by preserving native tissue structure and stromal complexity, allow spatially heterogeneous penetration and uptake of therapeutic agents. This feature provides a more stringent ex vivo platform to assess delivery efficiency, target engagement, and spatial distribution of miRNA-targeting agents under physical constraints that more closely resemble those of human tumors.98 In this setting, functionalized LNPs targeting tumor-specific or stroma-specific antigens represent a promising strategy to improve cell-type–selective delivery of miRNA therapeutics. In addition, labeling these LNPs enables direct visualization of nanoparticle localization and cargo delivery across distinct tumor and stromal compartments, providing insights into delivery efficiency and spatial constraints under physiologically relevant conditions. Finally, the integration of these ex vivo models with high-resolution analytical approaches would further enhance their translational value. PDO systems retaining endogenous immune populations have already been subjected to single-cell transcriptomic profiling to resolve tumor and immune heterogeneity at cellular resolution99 while spatially resolved transcriptomics has been successfully applied to intact human tumor tissues to define immune niches and predict response to immunotherapy.102 Hence, the integration of PDTFs-based platforms with spatially resolved transcriptomic and miRNA-specific in situ hybridization approaches (eg, miRNAscope) may enable spatially mapping of miRNA expression within preserved immune and stromal compartments. This would allow contextualizing miRNA localization within the transcriptional and functional landscape of the tumor and its immune microenvironment. Taken together, PDOs and PDTFs can be viewed as complementary platforms that can substantially improve the predictive value of miRNA-based immunomodulatory therapies. Conclusions and perspectives An expanding body of evidence underscores the pivotal role of miRNAs in orchestrating immune evasion and shaping the TIME, thereby contributing to resistance against ICIs. This review highlights the multifaceted mechanisms through which miRNAs suppress antitumor immunity and promote tumor progression. Targeting dysregulated miRNAs represents a promising therapeutic avenue to reprogram the TIME and enhance the efficacy of immunotherapy. Notably, miRNA-based therapeutics potentially offer a distinctive advantage over conventional protein-targeting approaches due to their ability to simultaneously modulate multiple layers of the immune response. This property is particularly relevant within the TIME, where immune evasion typically arises from complex and highly coordinated mechanisms rather than a single dominant pathway. So, incorporating miRNA therapeutics into immunotherapy regimens may unlock synergistic effects, particularly in tumors with immune-excluded or immune-suppressed microenvironments. Ultimately, developments in advanced platforms for preclinical studies may facilitate the translation of miRNA–ICI combination strategies into the clinic, driving progress in the field of precision immunotherapy.