Abstract
Non-coding RNAs (ncRNAs) regulate antitumor immunity but clinical use is hindered by instability, poor specificity, and immunogenic risk. Lipid nanoparticles (LNPs) offer a promising delivery solution due to biocompatibility, a modifiable surface, and efficient ncRNA encapsulation. Preclinical and early clinical studies have demonstrated the efficacy of LNP-ncRNA therapies in various cancers, including respiratory, digestive, and neurologic malignancies, and in cancer immunotherapy. This review evaluates engineered LNPs administered via different routes, explains the mechanisms that promote antitumor effects, and highlights advances in overcoming delivery barriers to enhance immunotherapy. Current limitations of engineered LNPs are also analyzed, essential translational challenges are identified, and future directions to facilitate the clinical application of LNP-based ncRNA therapies are suggested.
keywords
Introduction
Current mainstream immunotherapies, including immune checkpoint inhibitors (ICIs), chimeric antigen receptor (CAR)-T cells, cancer vaccines, and oncolytic viruses (OVs), have achieved significant efficacy in multiple cancer types, even enabling long-term survival in some advanced-stage patients1. RNA therapeutics present distinct advantages over conventional drugs by facilitating precise modulation of target protein expression. Moreover, RNA therapeutics are characterized by superior design flexibility and regulatory potential2. Among the RNA therapeutics, non-coding RNAs (ncRNAs), including long non-coding RNAs (lncRNAs), microRNAs (miRNAs), and circular RNAs (circRNAs), have emerged as promising therapeutic targets. These molecules regulate gene expression, immune checkpoint activity, and the tumor microenvironment (TME), thereby overcoming immune resistance and enhancing immunogenicity3.
However, the clinical translation of ncRNA-based therapies is hampered by delivery challenges. Specifically, unmodified ncRNAs are susceptible to degradation and demonstrate poor cellular uptake4. To enhance transfection efficiency, it is necessary to chemically modify ncRNAs, optimize delivery systems, or implement both strategies5. Lipid nanoparticles (LNPs) represent a crucial delivery platform, effectively protecting ncRNAs and improving cellular uptake and have been widely utilized in vaccine and therapeutic development6. Furthermore, an LNP-based targeting system facilitates the specific delivery of therapeutics to tumors, enabling efficient tumor clearance and precise TME modulation7. LNP-RNA complexes can also augment vaccine efficacy through adjuvant effects, stimulating antitumor immunity and suppressing tumor progression8. The global COVID-19 vaccines have validated LNPs as an efficient, scalable, and cost-effective delivery technology, although long-term storage and clinical translation issues persist.
This review provides a systematic analysis of recent advances in LNP-mediated ncRNA delivery for cancer immunotherapy. Design strategies, mechanisms of action, and practical applications of synergistic LNP-ncRNA systems are emphasized. The review further analyzes key challenges in clinical translation, including hepatic sequestration, immunogenicity, and scalability, while outlining future research directions. The aims of this review are to provide a scientific foundation and a developmental perspective for advancing the clinical translation of LNP-ncRNA combination therapies.
Cancer immunotherapy
Noteworthy progress in cancer immunotherapies has been achieved in recent years, including the clinical application of ICIs (Figure 1A), CAR-T cell therapy (Figure 1B), OVs (Figure 1C), cancer vaccines (Figure 1D), and NP-based strategies (Figure 1E), which have demonstrated promising antitumor efficacy, yet remain constrained by distinct limitations.
Principal mechanisms underlying cancer immunotherapy. This schematic illustrates five predominant therapeutic strategies that use distinct mechanisms to eradicate tumor cells. (A) Immune checkpoint inhibitors (ICIs) constitute a therapeutic approach utilizing mono-clonal antibodies to block interactions between inhibitory receptors on T cells and the corresponding ligands expressed on tumor cells or antigen-presenting cells (APCs). APCs present tumor-derived antigens via major histocompatibility complex (MHC) molecules on their surface, which are recognized by the T-cell receptor (TCR) and provide the initial activation signal for T cells. Immune checkpoint pathways mediate inhibitory effects. Specifically, the engagement of CTLA-4 on T cells with CD80/CD86 competitively impedes co-stimulatory signaling, whereas the binding of PD-1 on T cells to its ligand (PD-L1) on tumor cells or APCs suppresses T cell effector function. Therapeutic blockade using anti-CTLA-4 or anti-PD-1/PD-L1 mitigates these immunosuppressive signals within the TME, thereby restoring the antitumor effector functions of cytotoxic T lymphocytes. (B) CAR-T cell therapy involves the genetic modification of a patient’s autologous T cells to express CARs that specifically target TAAs. The canonical CAR structure consists of an extracellular antigen recognition, a transmembrane, and an intracellular signaling domain. Following ex vitro expansion, these engineered T cells are reinfused into the patient, generating an immune cell population capable of selectively identifying and efficiently eradicating tumor cells. (C) Oncolytic viruses (OVs) are genetically engineered to selectively replicate within and lyse tumor cells. Upon infection, OVs induce tumor cell lysis through viral replication, leading to the release of immunostimulatory molecules, such as tumor-associated antigens (TAAs) and damage-associated molecular patterns (DAMPs). These signaling molecules recruit dendritic cells (DCs) with GM-CSF further promoting DC recruitment and maturation. Activated DCs subsequently cross-present tumor antigens to CD8+ T cells, eliciting tumor-specific cytotoxic T lymphocyte responses. This cascade expands the localized cytolytic effect of viral infection into a systemic antitumor immune response. (D) Cancer vaccines are developed based on genomic analyses of tumor tissue, including WES and RNA-seq, to identify tumor neoantigens. These neoantigens are subsequently prioritized and optimized to develop DNA, RNA, or peptide vaccines. This process effectively primes and expands tumor-specific T cells, facilitating the infiltration into tumor sites and enhancing cytotoxic activity. (E) Engineered or activated nanoparticles (NPs) achieve targeted delivery by recognizing and binding specific cell surface receptors, such as G protein-coupled receptors, through active targeting mechanisms, followed by cellular internalization through endocytosis. Post-internalization, NPs are encapsulated within endosomes, subsequently undergoing endosomal escape and drug release, thereby enabling the intracellular delivery of immunomodulators or other therapeutic molecules. This strategy enables precise modulation of immune responses, thereby augmenting overall antitumor immunity. APCs, antigen-presenting cells; CAR, chimeric antigen receptor; CTLA-4, cytotoxic T-lymphocyte-associated protein 4; DAMPs, danger-associated molecular patterns; DCs, dendritic cells; GM-CSF, granulocyte-macrophage colony-stimulating factor; ICIs, immune checkpoint inhibitors; MHC, major histocompatibility complex; NPs, nanoparticles; OVs, oncolytic viruses; PD-1, programmed cell death protein 1; RNA-seq, RNA sequencing; TAAs, tumor-associated antigens; TCRs, T cell receptors; TME, tumor microenvironment; TSAs, tumor-specific antigens; WES, whole-exome sequencing. Figure created with BioRender.com.
ICIs
ICIs restore antitumor activity by blocking inhibitory signals on T cells and have been shown to have significant efficacy in various malignancies9. PD-1/PD-L1 inhibitors elicit durable responses and survival benefits as monotherapy or combined with radiotherapy or chemotherapy10. For example, a pH-responsive nanovaccine (dPEDE-A@M32) co-delivering a STING agonist and neoantigens augments immune responses when used with anti-PD-1 antibody11. Similarly, PD-1-IL2v expands tumor-reactive CD8+ T cells and promotes the TFH/TH1-like CD4+ T cell phenotype12, while a local delivery system (αPD-1@Lv/HPAGelα) inhibits tumor growth, metastasis, and recurrence, while establishing immune memory13. Combining cytotoxic T lymphocyte associate protein-4 (CTLA-4) and PD-1 blockers synergistically enhances the immune response, suppresses evasion, and induces durable memory14. An example is pembrolizumab with low-dose ipilimumab, which reinvigorates exhausted T cells and drives clonal expansion in melanoma patients refractory to PD-1/PD-L1 monotherapy15. Future immunotherapy development should focus on such engineered bispecifics and optimized sequential strategies to balance potent activation with controlled toxicity.
CAR-T cell therapy
CAR-T cell therapy combines T cell cytotoxicity with antibody-targeting specificity through engineered CARs. Application of CAR-T cell therapy in solid tumors faces challenges, although promising results have been reported with CT041, a CAR-T cell product targeting CLDN18.2 in gastric cancer16. Major limitations of CAR-T cell therapy are as follows: risk of secondary malignancies17; insufficient in vivo persistence and functional stability due to unstable transgene expression, host immune responses, and T-cell exhaustion18; and a scarcity of true tumor-specific antigens (TSAs) with most targets being tumor-associated antigens (TAAs) that carry the risk of off-target toxicity19. Innovative engineering strategies, such as modifying epidermal growth factor receptor variant III (EGFRvIII)-targeted CAR-T cells to incorporate transforming growth factor-β receptor variant II (TGF-βRII) and resist TGF-β-mediated suppression, are being developed to overcome these barriers20. Future development must prioritize optimized CAR designs to mitigate toxicity, novel TSA discovery, optimal dosing regimens, and TME-modulating combination strategies to expand CAR-T therapy toward safer, more effective personalized immunotherapy.
OVs
OVs selectively lyse tumor cells, inducing immunogenic cell death (ICD) and systemic antitumor immunity. The tumor tropism, immunogenicity, and safety of OVs can be enhanced through genetic engineering21. Combining OVs with ICIs (e.g., PD-1/PD-L1/CTLA-4 blockers) synergistically remodel the TME, showing enhanced efficacy in immunologically “cold” tumors22. The recombinant adenovirus (rAd) carrying the granulocyte-macrophage colony stimulating factor (GM-CSF) gene, named rAd.GM, combined with anti-PD-L1/CTLA-4 antibodies remodeled the TME, activated CD8+ T cells and memory T cells, suppressed Tregs and M2 macrophages, and inhibited tumor growth in triple-negative breast cancer (TNBC)23. Furthermore, OVs can synergize with CAR-T therapy by delivering immunomodulatory factors (e.g., neutralizing TGF-β) to enhance CAR-T cell infiltration and function24. It’s necessary to optimize OV design with novel targeting ligands and advance preclinical/clinical development of these rational combinations.
Cancer vaccines
Cancer vaccines elicit antigen-specific immune responses using antigens in diverse formats (whole cells, peptides, or nucleic acids). The optimal design for a cancer vaccine must overcome immunosuppression and activate dual immune recognition pathways25. Although recent trials have shown improved relapse-free survival (RFS) in patients with melanoma26 and pancreatic cancer (PC)27, most patients still face limited efficacy challenges28.
NP-based immunotherapy
Nanoscale drug delivery systems (NDDSs) enhance immunotherapy efficacy and reduce systemic toxicity through improved stability, biocompatibility, and targeting29. For example, polymer-coated mesoporous silica NPs co-delivering volasertib and anti-PD-L1 achieved a 5-fold dose reduction with superior antitumor effects30. Other NPs, such as yolk-shell structures inducing ICD and downregulating PD-L131 or hybrid NPs combining photothermal therapy (PTT) with M2 macrophage depletion to activate T-cells32, have been shown to have significant potential across cancers. When combined with ICIs, these strategies produce pronounced synergy in models, yet clinical translation challenges persist.
LNP-mediated immunotherapy
LNPs constitute a transformative platform for drugs, genes, and vaccines, and are valued for excellent biocompatibility, scalable production, high drug-loading capacity, and potential for multifunctional modifications33. Preclinically, innovative LNP designs show significant efficacy against respiratory, digestive, and neurologic diseases (Table 1), while LNP-based cancer therapies have advanced into clinical trials (Table 2).
Preclinical evaluation of LNP-based therapeutics for multi-system diseases
Advances in LNP-based clinical trials in cancer
In contrast, metal-organic frameworks (MOFs), although structurally tunable, remain underexplored for ncRNA delivery with degradation kinetics requiring further investigation49. Dendrimers enable facile multi-functionalization but face clinical limitations due to potential biotoxicity from cationic charge and complex synthesis50. Overall, LNPs currently demonstrate superior performance and hold greater translational potential for ncRNA delivery.
LNP structure
LNP is the cornerstone of non-viral gene delivery platforms, the performance of which relies on the optimized composition of ionizable cationic lipids, helper phospholipids, polyethylene glycol (PEG)ylated lipids, and cholesterol51. Ionizable lipids are fundamental components of LNPs because protonation in acidic endosomes enhances encapsulation and induces membrane destabilization for cytoplasmic release52. Cholesterol stabilizes LNP structure and can be modified to redirect biodistribution, although the ratio is critical to prevent crystallization53. Advanced ionizable lipids need less helper phospholipid but a minimum quantity ensures stable encapsulation54. PEG-lipids modulate particle properties [optimal size (~140 nm) and PEG content (~1.5 mol%)]55, circulation time, organ targeting (including liver tropism via ApoE)56, and brain delivery in models57.
The distinct TME enables engineered LNPs for localized, triggered release. Key strategies include redox-sensitive linkages [cleaved by high glutathione (GSH)]58, enzyme-responsive systems [e.g., matrix metalloproteinases (MMPs)]59, and ATP-responsive platforms for nucleic acid release60. LNPs an also be actively targeted to immune cells via surface ligands, such as CD4 antibody for T-cells61 or single-chain variable fragment (scFv) for dendritic cells62. LNPs also serve as potent vaccine adjuvants, which was validated in SARS-CoV-2 vaccines63. However, the immunostimulatory properties require careful safety optimization to mitigate risks, like aberrant immune activation, complement response, and thrombosis. Future work must balance immunostimulation with biocompatibility to advance safer clinical translation.
LNP in respiratory system cancers
LNPs demonstrate preclinical promise for respiratory cancer immunotherapy via inhalation or intravenous delivery64. Strategies to enhance aerosol stability are critical, such as a charge-assisted stabilization (CAS) strategy using peptide-lipid conjugates35 and optimized spray-drying to produce effective LNP powders65. Replacing PEG-lipids with zwitterionic ionic polymers (ZIP) also improve stability and expression post-nebulization, indicating good tolerance36.
A key advance is lung-selective organ targeting (Lung SORT) technology. This platform enables efficient delivery of CRISPR-Cas9 nucleases and adenine base editors (ABEs) to all major lung cell types66. This technology demonstrated significant gene therapy potential and durable efficacy in a cystic fibrosis (CF) model34,37. Similarly, siRNA SORT LNPs achieved 88% target knockdown with high lung selectivity38. Cationic lipid/polymer (CLP) strategies further enhance lung tropism67. However, cationic surfaces pose a thrombosis risk by potentially activating coagulation68, necessitating systematic biocompatibility evaluation. Emerging technologies, such as gas therapy LNPs for nitric oxide release69 and magnesium stabilizers for vaccines70, provide preclinical proof-of-concept for combinations. Future efforts must optimize formulations to reduce thrombosis risk while integrating gene editing and RNA interference (RNAi) for safer, personalized respiratory tumor interventions.
LNP in digestive system cancers
LNPs offer innovative solutions to overcome the unique barriers of intestinal tract for immunotherapy. While mucosal routes (e.g., oral or rectal) are constrained by mucus barriers and carrier instability71, advanced LNP designs show promise.
Pancreas-targeting LNPs (Pantgt LNPs) deliver interleukin-12 (IL-12) mRNA via intraperitoneal injection for pancreatic adenocarcinoma (PAAD), activating immune-related genes, remodeling the immunosuppressive microenvironment (ISME), and achieving complete tumor regression72. Other strategies include Tween 80-based LNPs with tLyp-1 peptide for siKRAS-LNA delivery39, a dual-targeting anti-MSLN nanobody system for gemcitabine delivery40, and novel LHHK-LNPs for immunomodulator delivery without pancreatic injury73. LNPs enable local mucosal immunity in colorectal cancer (CRC), where immune cell infiltration is inefficient74. Intramuscular LNP-mRNA cancer vaccines with all-trans retinoic acid (ATRA)41 and oral LR-S-CD/CpG@LNPs42 suppress tumor growth in models with oral delivery also modulating gut microbiota75. LNPs also deliver small interference RNAs (siRNAs) for gastric/liver cancers43,44. Future efforts should optimize LNPs with mucoadhesive or gut-targeting ligands for better mucus penetration, design condition-responsive release, and explore combination delivery routes to activate systemic and local mucosal immunity.
LNPs in central nervous system tumors
The blood-brain barrier (BBB) is a specialized vascular structure that protects the brain but prevents nearly all therapeutics from entering the brain tumors76, which must be overcome for therapy. Early strategies used invasive direct injection of LNPs, demonstrating efficacy but carrying risks77. Intravenous injection offers better compliance78, driving preclinical research into non-invasive delivery systems. Current strategies focus on surface-functionalized LNPs with targeting ligands (e.g., T7 peptides or monoclonal antibodies) to exploit receptor-mediated transcytosis across the BBB45. Other innovative designs integrate targeting, penetration enhancement, and stimulus-responsive release79, in which novel ionizable lipids can disrupt endothelial tight junctions and enhance permeability46,47.
In addition to chemical modification, physical methods, like focused ultrasound with microbubbles, can temporarily open the BBB for LNP entry48. Intranasal administration presents an alternative route, bypassing first-pass metabolism and enhancing brain localization80. Future efforts should integrate optimized lipid compositions, targeted ligands, and physical methods for efficient BBB penetration. Concurrently, improving long-term safety, reducing immunogenicity, and advancing personalized dosing in preclinical studies are essential to expand treatment options for brain tumors and other neurological disorders.
LNP-mediated ncRNAs in cancer immunotherapy
In recent years, LNP-delivered ncRNAs have demonstrated tremendous potential to remodel TME in cancer immunotherapy (Figure 2).
LNP-ncRNAs remodel the tumor microenvironment (TME). The TME is remodeled through the synergistic actions of three principal mechanisms, which collectively enhance antitumor immune responses. (A) Targeting MDSCs: ncRNAs mitigate the immunosuppressive properties of MDSCs by reducing the population and impairing functional capabilities. This process is achieved via multiple pathways, including the inhibition of MDSC development and differentiation to reduce the number, attenuation of immunosuppressive activities, and blockade of the recruitment to tumor sites. Notably, MDSCs express elevated levels of PD-L1 on the surface, which interacts with PD-1 receptors on T cells, transmitting inhibitory signals that induce T cell exhaustion. Interventions mediated by ncRNAs alleviate MDSC-driven immunosuppression, restore T cell cytotoxicity, and thereby enhance tumor cell eradication. (B) Activating CD8+ T cells: the activation of CD8+ T cells proceeds via two primary pathways: the indirect pathway involves the alleviation of MDSC-mediated suppression, thereby reinstating the cytolytic function of T cells. The direct pathway utilizes LNPs modified with anti-CD8 antibodies to deliver CAR-encoding mRNA specifically to CD8+ T cells. This targeted delivery facilitates endogenous CAR expression, converting these cells into CAR-T cells capable of recognizing TAAs and exerting potent cytotoxic effects. (C) Reprogramming TAMs: ncRNAs promote the phenotypic shift of TAMs from a pro-tumorigenic M2 state to an anti-tumorigenic M1 state. M2 TAMs are characterized by high PD-L1 surface expression, which engages PD-1 on T cells to suppress their activity, whereas M1 TAMs exhibit low PD-L1 expression and activate T cells through the secretion of cytokines, such as IL-12. This phenotypic reversion alleviates TAM-mediated immunosuppression, restores T cell cytotoxic function, and subsequently enhances tumor cell elimination. Additional strategies include inhibiting TAM recruitment to the TME and modulating the pro-tumorigenic functions to reduce the immunosuppressive impact on immune responses. IL-12, interleukin-12; LNPs, lipid nanoparticles; MDSCs, myeloid-derived suppressor cells; ROS, reactive oxygen species; TAMs, tumor-associated macrophages; TME, tumor microenvironment. Figure created with BioRender.com.
LNP-ncRNAs in solid tumors
LNP-mediated siRNAs
LNP-siRNA technology offers a transformative approach to overcome tumor resistance to immunotherapy by enabling precise, sequence-specific silencing of immune-related genes (Figure 3A). Following cytoplasmic delivery via endosomal escape, siRNA is incorporated into the RNA-induced silencing complex (RISC) with Argonaute 2 (AGO2), which uses sequence complementarity to cleave mRNA transcripts encoding immune checkpoint proteins, like PD-1 or CTLA-4. This post-transcriptional gene silencing alleviates T inhibition81. The platform effectively reprograms the ISME, promotes activation of T cells, NK cells, and M1-type macrophages, and exhibits strong synergistic effects with ICIs, agonist antibodies, or cancer vaccines.
LNP-mediated ncRNAs in cancer immunotherapy. LNPs encapsulate siRNA, circRNA, miRNA, or lncRNA to form stable LNP-ncRNA complexes. Upon intravenous administration, these complexes enter the systemic circulation. Through surface modification with targeting ligands that specifically recognize receptors on target cells, the complexes achieve active targeting and are subsequently internalized via receptor-mediated endocytosis, resulting in sequestration within endosomes. Within the acidic endosomal environment, the ionizable lipids in the LNPs become protonated, disrupting the endosomal membrane and thereby facilitating the efficient release of the ncRNA cargo into the cytoplasm. Once released, ncRNAs modulate immune responses by regulating gene expression through several distinct pathways: (A) siRNA: siRNA is incorporated into the RISC and exhibits perfect complementarity to the target mRNA, enabling precise cleavage and degradation of the transcript, which results in highly specific gene silencing. (B) circRNA: circRNA primarily acts as a ceRNA by sequestering intracellular miRNA via a “molecular sponge” mechanism, thereby preventing miRNA from repressing the endogenous mRNA target and leading to the upregulation of a tumor suppressor (e.g., immune-related gene). In addition, some circRNAs containing IRES can be translated into small peptides that modulate immune signaling pathways. (C) miRNA: miRNAs, either as mimics or inhibitors, are incorporated into the RISC and interact with AGO proteins. By binding with partial complementarity to the 3′-UTR of the target mRNAs, resulting in translational repression or mRNA degradation and consequent downregulation of gene expression. (D) lncRNA: lncRNA acts as a molecular regulator by binding specific miRNA through complementary base pairing, thereby sequestering the miRNA and preventing it from binding to and promoting degradation of its target mRNA. 3′-UTR, 3′-untranslated region; AGO, argonaute; ceRNA, competing endogenous RNA; IRES, internal ribosome entry sites; RISC, RNA-induced silencing complex. Figure created with BioRender.com.
Targeting RNA modifications is a key focus. N6-methyladenosine (m6A) regulates RNA metabolism and influences tumor progression and TME. The cap-adjacent modification, m6Am, also enhances transcript stability and promotes therapy resistance82. LNP-siYthdf1 targeting the m6A reader protein, YTHDF1, suppressed the EZH2–IL-6–myeloid-derived suppressor cell (MDSC) axis, reduced of MDSC infiltration, restored T-cell function, and inhibited tumor progression in a preclinical non-alcoholic steatohepatitis-associated hepatocellular carcinoma (NASH-HCC) model, 43. Similarly, LNP-CMsiRNA against the m6Am methyltransferase, PCIF1, downregulated FOS mRNA stability, inhibited TGF-β signaling, remodeled the ISME, and promoted NK cell infiltration in CRC83. Knockdown of Ythdf1 or Pcif1 has been shown to sensitize tumors to anti-PD-1 therapy in models.
Tumor-associated macrophages (TAMs), often the pro-tumor M2 phenotype, are another major target84. The cationic nanoparticle, CL4H6, silenced signal transducer and activator of transcription 3 (STAT3) and hypoxia inducible factor 1α (HIF-1α) in HCC, promoting infiltration of CD11b+ macrophages and enriching CD169+ M1-type macrophages, thereby reversing TAM pro-tumor functions with significant increases in immunostimulatory agents85. To overcome ICI resistance, innovative strategies, like genetically engineered cell membrane vesicles (GECMs), are being explored. LNPs coated with PD1-overexpressing membranes (LNP@mPD1) blocked PD-1/PD-L1 and achieved tumor targeting in preclinical studies. Loaded with siAdar1, this platform suppresses adenosine deaminase acting on RNA 1 (ADAR1) expression, promoted IFN-β/γ production, sensitized cells to IFN-γ, inhibited tumor growth, and enabled 50% survival beyond 60 days86.
The BBB challenges central nervous system drug delivery. Cationic lipids can enhance BBB permeability. The novel lipid, BAMPA-O16B, improved siRNA delivery to glioblastoma (GBM) in preclinical models. The BAMPA-O16B/siCD47/siPD-L1 system crossed the BBB and targeted GBM cells with 1.78-fold increased delivery, showing synergistic antitumor effects via dual CD47 and PD-L1 downregulation47. Similarly, LNPs with siSIRPα reduced SIRPα expression by ~50% in macrophages, disrupted CD47-SIRPα interactions, sensitized ovarian cancer (OC) to chemotherapy, and suppressed invasive behavior87.
Combination strategies synergize ICIs with immune agonists88. Intratumoral injection allows high local concentration. Preclinically, intratumoral LNPs co-loaded with pDNA and siRNA enhanced immune activation in B16F10 tumor-bearing mice, reducing tumor size by ~50% (low-dose) and ~80% (high-dose), inducing regression and immunologic memory89. A phase I or II trial (NCT0373731) is evaluating intratumoral mRNA-2752 LNP-encoding human OX40L, IL-23, and IL-36γ with or without durvalumab. Another preclinical system used dual LNPs with VISTA siRNA and TLR9 agonist CpG in murine tumors, enabling dual modulation, promoting DC/macrophage activation, achieving 83% tumor clearance, and minimizing systemic inflammation90. Combination PD-L1 blockade with tumor neoantigen vaccines is promising for cancers, like non-small cell lung cancer (NSCLC) with driver mutations (e.g., EGFR)91. Preclinical studies reported that co-delivery of PD-L1 siRNA and an EGFR-derived peptide via cationic PEI-LNPs elevated IFN-γ and TNF-α, reduced IL-10, enabled sustained release of immunostimulatory factors, remodeled the ISME, and induced durable antitumor immune responses, achieving a 76.2% apoptosis rate in lung cancer (LC) cells92.
In summary, LNP-siRNA technology reprograms the TME through precise gene silencing, promotes antitumor immunity, and reverses immune tolerance. It shows marked synergy with ICIs, agonists, chemotherapeutics, and vaccines, enhancing outcomes. Numerous LNP-siRNA formulations have shown potential in overcoming immunotherapy resistance in preclinical and early clinical studies. Future work should optimize target selection and delivery efficiency, advance personalized combination strategies, and further explore the platform potential against tumor heterogeneity and metastasis.
LNP-loaded circRNAs
CircRNAs possess superior chemical stability and RNase R resistance due to a covalently closed structure93. CircRNAs function primarily as competing endogenous RNAs (ceRNAs), sequestering miRNAs via a molecular “sponge” mechanism to upregulate tumor suppressor and immune-related genes. Some circRNAs containing internal ribosome entry sites (IRES) can also be translated into immunomodulatory peptides (Figure 3B). Dysregulation is a key oncogenic driver. For example, LNP-delivered circUGP2 sponges (miR-3191-5p) to upregulate ADGRB1, inhibiting MDM2-mediated p53 degradation and suppressing intrahepatic cholangiocarcinoma (ICC) progression94.
circRNA platforms enable in vivo antigen expression in cancer vaccination, activating antigen-presenting cell (APC) and inducing tumor-specific T-cell responses95. Preclinically, the LNP-circRNA vaccine significantly increased cytotoxic T cells, elevated IFN-γ and TNF-α, and suppressed tumor growth. In combination with adoptive T-cell therapy, the LNP-circRNA induced complete tumor regression and prolonged survival in advanced melanoma96.
For cytokine delivery, LNP-delivered IL-12 circRNA enables sustained local expression, eliciting durable antitumor immunity. Combined with anti-PD-L1, LNP-delivered IL-12 circRNA induces significant tumor regression. Optimized LNPs, like H11A1B3-LNP, show a four-fold higher circRNA transfection efficiency than clinical benchmarks97. Furthermore, the circ-arRNA platform achieves efficient in vivo RNA editing, successfully repairing a TP53 mutation with 48.48% efficiency and sensitizing tumors to chemotherapy in TNBC98. Synergy with siRNA strategies enhances immune modulation. Silencing oncogenic circAKT3 via LNP-delivered siRNA inhibits prostate cancer (PCa) progression99. Advanced targeting is achieved with like, AS1411 aptamer. AS1411-modified LNPs delivering si-circPDHK1 inhibit tumor proliferation in clear cell renal cell carcinoma (ccRCC) and show enhanced antitumor efficacy100.
circRNAs offer considerable potential in therapy and immune regulation due to stability and multifunctionality. circRNAs form a promising foundation for next-generation RNA combination therapies as vaccines, editing templates, or in combination with siRNA.
LNP-encapsulated miRNAs
miRNAs are key cytoplasmic negative regulators. miRNA mimics or inhibitors incorporate into the RISC and bind to target the mRNA 3′-untranslated region (3′-UTR) with partial complementarity, leading to mRNA degradation or translational repression (Figure 3C). Dysregulation of miRNA mimics or inhibitors significantly influences cancer progression. For example, inhibiting the oncomiR (miR-503) promotes apoptosis and suppresses tumor growth in multiple myeloma (MM)101. Preclinical research has shown that miR-155 enhances antitumor immunity by promoting pro-inflammatory cytokines (e.g., TNF-α and IL-6) and inducing M1 macrophage polarization102. LNP-encapsulated miR-155 Cy5 inhibitor suppresses the β-catenin/TCF4 pathway, reduces intracellular Cu2+, and impedes tumor cell proliferation103.
Clinical translation of miRNA therapeutics is hindered by inherent instability and poor delivery given that the SIRPα-CD47 axis facilitates immune evasion104. Anti-SIRPα antibody-modified LNPs (LNP-miR155@aSIRPα) specifically target TAMs in melanoma models. This system blocks the SIRPα-CD47 interaction and delivers miR-155 intracellularly, reprogramming TAMs to an antitumor phenotype, increasing pro-inflammatory secretion and phagocytic activity, resulting in a 60% survival rate105. Hyaluronic acid (HA)-modified LNPs target CD44-overepressing glioma cells106. Studies demonstrate that miR-181a-loaded HA-LNPs maintain high transfection efficiency and show therapeutic potential in GBM models107.
In addition to cytoplasmic roles, miRNAs function in nuclear epigenetics, including super-enhancers108. The miR-200c precursor is a nuclear-activated miRNA (NamiRNA) within a super-enhancer in PC. Activation upregulates PTPN6, which suppresses tumor proliferation, and LNP-miR-200c treatment inhibits hepatic metastasis109.
Current miRNA research has evolved into multidimensional systems integrating targeted delivery, immune regulation, and epigenetic editing. Functional LNP platforms enabling dual targeting and stimulus-responsive release are advancing the precision of cancer therapy.
LNP-delivered lncRNAs
lncRNAs are non-coding transcripts >200 nucleotides in length that regulate cellular functions110, including the TME, where lncRNAs influence T lymphocyte activation. LNP-based delivery systems enable precise targeting of such lncRNAs to inhibit the PD-1/PD-L1 pathway and restore T cell-mediated tumor cytotoxicity111. Some lncRNAs also encode functional peptides. The SRSP peptide from lncRNA LOC90024 is pro-tumorigenic112, while lnc-AP acts as a tumor suppressor in CRC113.
Targeting lncRNAs with antisense oligonucleotides (ASOs) is a promising strategy. ASOs, mediate RNase H-dependent degradation of target lncRNAs and LNPs facilitate efficient in vivo delivery (Figure 3D). For example, PEI@VPS LNPs deliver ASO against VPS9D1 in CRC reduce metastasis and enhance ICI efficacy114. Similarly, LNP-mediated delivery of ASOs targeting the lncRNAs, HOTAIR and MALAT1, suppress tumor growth and metastasis in breast cancer (BC) models115.
LNP technology overcomes the instability and delivery challenges of lncRNA therapeutics. Future work should optimize LNP targeting specificity and elucidate lncRNA regulatory networks to develop safer, more precise lncRNA-based combination immunotherapies.
LNP-ncRNAs in hematologic malignancies
Recent LNP technologies advance cell-specific ncRNA delivery to hematopoietic tissues, although research has only been preclinical. A key challenge in acute myeloid leukemia (AML) models is eliminating chemotherapy-resistant leukemia stem cells (LSCs)116. To overcome limitations of in vivo CRISPR-Cas9, such as suboptimal efficiency and off-target effects, a biomimetic strategy was developed. This system co-loads LNP-encapsulated Cas9/sgRNA ribonucleoprotein (RNP) and the chemokine, CXCL12α, onto mesenchymal stem cell membrane-coated nanofibers (MSCM-NFs). Upon bone marrow injection, it exploits the CXCL12α/CXCR4 axis to recruit leukemia cells, enabling highly efficient, localized gene editing117.
CD3 antibody-modified LNPs deliver optimized plasmids encoding a CAR gene and IL-6 shRNA in CAR-T cell therapy. Intravenous administration generates IL-6-downregulated CAR-T cells in vivo, which enhances antitumor immunity118. RNA activation (RNAa), mediated by small double-stranded RNAs (dsRNAs), offers a novel epigenetic strategy by enhancing endogenous gene transcription119. Specifically, sgRNA targeting CDH13 can overcome BCR-ABL1-independent resistance by inhibiting NF-κB signaling and inducing apoptosis120. Taken together, LNP-ncRNA strategies have demonstrated promising therapeutic efficacy in diverse preclinical cancer models (Table 3).
Therapeutic efficacy of LNP-ncRNA strategies in preclinical cancer models
Combined strategies that integrate epigenetic modulation and gene editing show promise against drug resistance. Given the distinct functions, delivery strategies, and translational challenges of various ncRNAs (Table 4), precise design of a personalized LNP system is essential.
Comparison of siRNA, miRNA, circRNA, and lncRNA in immunotherapy
Current challenges and future directions
Although LNP-ncRNAs demonstrate considerable potential in cancer immunotherapy, clinical translation faces a series of critical challenges, including but not limited to hepatic sequestration, thrombosis risk associated with cationic lipids, repeat-dose toxicity, cytokine release, long-term biodistribution characteristics of ncRNAs, as well as scalable manufacturing processes and storage stability.
Targeting challenges and solutions
Intravenously administered LNPs exhibit pronounced hepatic tropism due to surface adsorption of ApoE and subsequent low-density lipoprotein (LDL) receptor-mediated endocytosis. This effect results in non-specific accumulation of 30%–90% of the administered dose in the liver121. Novel LNP platforms are being developed to address this challenge. For example, a siloxane-based ionizable lipid library enabled the creation of siLNPs, which achieved organ-specific ncRNA delivery to the livers, lungs, and spleens in mice122. In addition, biomimetic NPs, such as cell-membrane-coated NPs (CMCNs) and extracellular vesicles (EVs), leverage endogenous biological signals present on the surfaces to offer distinct advantages, including enhanced targeting, prolonged circulation, and reduced immunogenicity123. By fusing with the cell membranes derived from different sources, CMCNs can be functionalized for specific purposes, such as using cancer cell membranes to achieve homotypic targeting124. Immune cell membranes promote endothelial adhesion and extend circulation time125. Erythrocyte membranes facilitate immune evasion and long-term circulation126. Furthermore, novel ionizable lipids, such as those lipids derived from vitamin B5, have been developed and shown to specifically target lymphoid tissues, providing a new strategy for achieving tissue-specific delivery127.
Safety and toxicity concerns
The pulmonary-targeted design of LNPs, owing to the cationic nature, may trigger local coagulation and complement system activation, which can be further amplified through immunothrombotic mechanisms, such as neutrophil extracellular trap (NET) release and tissue factor exposure, ultimately exacerbating microvascular occlusion and lung injury68. To address this effect, ionizable lipids with finely tunable pKa values can be used, thereby effectively reducing non-specific adsorption to and activation of blood cells and endothelial cells128. In addition, co-delivery of anticoagulant agents with LNPs or pre-administration of low-dose heparin to neutralize negative charges on the vascular wall can effectively suppress coagulation initiation induced by cationic carriers129. Furthermore, when LNPs carry nucleic acids into the body, LNPs may be recognized by the immune system as virus-like particles, leading to immune cell activation and subsequent release of defensive cytokines130. In this context, low-dose regimens should be considered a key strategy, even during early-stage validation in both in vitro and in vivo bioassays. Moreover, larger NPs with complex surface modifications tend to induce stronger cytokine responses compared to smaller NPs with simpler surface chemistry131.
Manufacturing and scalability
Although LNP formation is a one-step self-assembly process, scaling the process to robust, GMP-compliant industrial manufacturing remains difficult. The key lies in optimizing process parameters to ensure batch-to-batch consistency and high yield. Studies have shown that dynamically controlling LNP formation and physicochemical properties is achievable by modulating solvent composition and mixing under specific melting conditions, followed by timed aqueous buffer injection132. Moreover, an aqueous scalable process suitable for mRNA vaccines has been developed, which is readily adaptable to GMP standards and supported by established characterization and quality control methods133. Further optimization of downstream tangential flow and sterile filtration will enhance the robustness and scalability of these manufacturing platforms134.
Storage stability issues
Storage stability is another critical barrier. Conventional LNP-ncRNA formulations often require storage at −70°C or −20°C, imposing substantial cold-chain challenges. Studies have demonstrated that blank NPs in 10% sucrose/PBS can retain in vivo potency after 30 d at −20°C. Kim et al.135 further confirmed that lyophilized and reconstituted mRNA-LNPs maintain structural integrity and bioactivity with stability exhibiting clear temperature dependence from −80°C to 4°C. Compared to the liquid formulations of Moderna and Pfizer/BioNTech COVID-19 vaccines, which are stable for only hours, lyophilized versions extended shelf-life to weeks, reducing reliance on ultra-cold chains136. Notably, novel ionizable lipids, like TOT-28, enable LNPs to retain stable for up to 8 weeks at 4°C or 25°C, offering a new strategy to overcome storage limitations137.
Clinical pipeline and future directors
The clinical translation of LNP-ncRNA therapies has significantly accelerated, as validated by the successful deployment of the BioNTech COVID-19 vaccine (BNT162b2)138. Looking forward, a diverse array of ncRNA-based therapies aimed at different pathways are currently undergoing clinical evaluation. For example, an siRNA therapy directed against EphA2 (NCT01591356) is presently in the phase assessing safety and tolerability. Concurrently, the dual inhibitor, STP705, which targets TGF-β1 and cyclooxygenase-2 (COX-2), has received approval for cholangiocarcinoma and non-melanoma skin cancer (NMSC). Furthermore, several RNA therapies utilizing LNP delivery systems are transitioning from preclinical investigations to early-phase clinical trials, including an siRNA therapy targeting the KRAS G12D mutation (NCT03608631) for PC and NBF-006 (NCT03819387), which is currently in phase I trials for NSCLC, PC, and CRC. In phase II clinical trials, STP705 (NCT0484483) is being evaluated for squamous cell carcinoma (SCC), while MTL-CEBPA (NCT04710641), a self-amplifying RNA (saRNA) therapeutic, is under investigation in patients with HCC. LNP-ncRNA therapies surge ahead, spanning COVID success to cancer trials, poised to redefine precision medicine.
Despite the persistence of numerous challenges in clinical translation, significant preclinical advances are steadily propelling this field toward clinical implementation. The successful transition is fundamentally contingent upon ensuring the safety, targeting specificity, and process optimization of novel therapeutics and the associated delivery systems. Machine learning (ML) supports personalized treatment strategies by integrating heterogeneous data from multiple sources, identifying key biomarkers, and developing predictive models139. Concurrently, modular platforms, such as POST, offer a systematic design framework for the design of LNP surface modifications, thereby enabling organ-specific delivery140. Collectively, these methodologies are instrumental in advancing the translation of LNP-ncRNA therapies from fundamental research toward clinical application.
Conclusions
LNP-ncRNA delivery has evolved into a programmable immunomodulation platform. The core advantage lies in a unique blend of precision and comprehensiveness. LNP-ncRNA enables sequence-specific genetic intervention against a wider array of targets than antibodies or small molecules, while the capacity for co-delivering diverse ncRNAs allows for synergistic remodeling of the TME beyond the reach of single-function mRNA therapies. To successfully harness this potential, translation should be guided by integrated principles of spatiotemporal precision, which is achieved through the following: active targeting and TME-responsive designs to ensure accurate delivery; synergistic breadth, enabled by rational ncRNA sequences and strategic combinations with conventional therapies; and translational robustness, built upon stable formulations, scalable manufacturing, and adaptable dosing regimens.
Realizing this future follows a logical clinical pathway. The near-term focus must be on validating safety and proof-of-concept for combination regimens in refractory cancers. Subsequently, mid-term efforts should concentrate on breakthroughs in organ- and cell-selective delivery to overcome fundamental distribution challenges. The long-term vision is the seamless integration of this platform into precision oncology, shifting from broad-spectrum applications to truly personalized therapeutic programming. Ultimately, translating the transformative promise of LNP-ncRNA immunotherapy into tangible patient outcomes necessitates the deep and sustained convergence of material science, immunology, and clinical medicine to reshape the cancer treatment landscape.
Conflict of interest statement
No potential conflicts of interest are disclosed.
Author contributions
Conceived and designed the analysis: Yufei Sheng, Zhaohui Gong.
Collected and performed the analysis: Yufei Sheng, Lulu Yang, Boyang Wang, Wentao Hu, Chengwei Zhou.
Wrote the paper: Yufei Sheng.
Revised the paper: Zhaohui Gong.
Acknowledgments
The authors thank their respective laboratory members and collaborators for critical review of this article. The authors apologize that space constraints prevent them from citing all relevant publications.
- Received December 30, 2025.
- Accepted March 2, 2026.
- Copyright: © 2026, The Authors
This work is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License.
References
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.↵
- 94.↵
- 95.↵
- 96.↵
- 97.↵
- 98.↵
- 99.↵
- 100.↵
- 101.↵
- 102.↵
- 103.↵
- 104.↵
- 105.↵
- 106.↵
- 107.↵
- 108.↵
- 109.↵
- 110.↵
- 111.↵
- 112.↵
- 113.↵
- 114.↵
- 115.↵
- 116.↵
- 117.↵
- 118.↵
- 119.↵
- 120.↵
- 121.↵
- 122.↵
- 123.↵
- 124.↵
- 125.↵
- 126.↵
- 127.↵
- 128.↵
- 129.↵
- 130.↵
- 131.↵
- 132.↵
- 133.↵
- 134.↵
- 135.↵
- 136.↵
- 137.↵
- 138.↵
- 139.↵
- 140.↵
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.