Nanocarrier-based delivery of siRNA therapeutics in rheumatoid arthritis: immune mechanisms and translational perspectives

Rheumatoid arthritis (RA) is a chronic systemic autoimmune disease characterized by persistent synovial inflammation, pannus formation, and progressive joint destruction. The pathogenesis of RA involves complex interactions between innate and adaptive immune responses, pro-inflammatory cytokine networks, and aberrant activation of intracellular signaling pathways, ultimately leading to irreversible cartilage and bone damage (1, 2). Despite advances in pharmacotherapy over the past decades, RA remains a major cause of disability and reduced quality of life worldwide (3).

Conventional therapeutic strategies, including methotrexate (MTX), nonsteroidal anti-inflammatory drugs (NSAIDs), and biologic agents targeting cytokines or immune cells, have greatly improved disease outcomes (4). However, these agents are often limited by incomplete efficacy, systemic adverse effects, high costs, and the emergence of drug resistance (5). These limitations highlight the unmet need for novel therapeutic approaches that can selectively and sustainably modulate disease-driving molecular mechanisms.

Nevertheless, the clinical translation of siRNA therapeutics faces significant challenges. Naked siRNAs are inherently unstable in biological fluids, prone to rapid enzymatic degradation, and exhibit poor cellular uptake and suboptimal biodistribution (6). Moreover, systemic administration can trigger off-target effects and unintended immune responses (7). These barriers underscore the critical need for effective delivery platforms capable of protecting siRNA, enhancing cellular internalization, and achieving site-specific release within the inflamed synovium.

Rheumatoid arthritis is a fundamentally multifactorial disorder driven by the interplay of numerous cytokines, multiple pathogenic cell subsets, and a constellation of intracellular signaling pathways including NF-κB, JAK/STAT, MAPK, and PI3K/Akt (1). Unlike small-molecule drugs or biologics that typically inhibit a single receptor or cytokine, siRNA therapeutics operate at the post-transcriptional level, enabling precise silencing of virtually any gene implicated in RA pathogenesis (6). This makes siRNA uniquely suited for a disease in which targeting a single cytokine or pathway often provides incomplete or transient benefit. Furthermore, multiple siRNA sequences can be combined within a single formulation, allowing simultaneous suppression of converging inflammatory mediators—such as TNF-α, IL-1β, or STAT3—to disrupt entire pathogenic circuits rather than isolated nodes (8). siRNA also provides the flexibility to modulate intracellular factors that are undruggable by conventional therapeutics, including transcription factors, adaptor proteins, or intracellular kinases that play central roles in synovial hyperplasia and immune dysregulation. Importantly, by selectively reprogramming macrophage polarization or fibroblast phenotypes, siRNA can reshape the RA microenvironment at a systems-level scale, addressing both inflammatory and structural components of disease. Together, these features position siRNA as a uniquely powerful modality for confronting the molecular complexity and cellular heterogeneity that define rheumatoid arthritis.

Nanocarrier-based drug delivery systems offer a promising solution to these challenges. Engineered nanomaterials—including lipid nanoparticles, polymeric nanocarriers, and inorganic nanoparticles—have shown the ability to encapsulate siRNA, protect it from degradation, and facilitate targeted delivery to diseased tissues (9 –11). Furthermore, advanced designs incorporating pH-, enzyme-, or reactive oxygen species (ROS)-responsive elements enable controlled and stimuli-sensitive release within the inflammatory microenvironment of RA (12). These innovations not only improve therapeutic efficacy but also minimize systemic toxicity, paving the way for the application of siRNA-based precision medicine in RA management.

Though several reviews have discussed siRNA therapeutics or nanocarrier platforms independently, an integrated analysis linking nanocarrier design principles to the immunopathology of rheumatoid arthritis is still lacking. Moreover, recent advances in dual-responsive polymeric systems, metal–organic framework (MOF)-based co-delivery platforms, macrophage-reprogramming nanocarriers, and lipid–polymer hybrid architectures have not been synthesized in previous literature. To address this gap, the present review provides a comprehensive and updated framework that connects nanomaterial engineering with immune-regulatory mechanisms, including cytokine modulation, macrophage polarization, angiogenesis, and intracellular signaling pathways. In addition, we incorporate a translational perspective, discussing protein corona formation, complement activation–related immunotoxicity, GMP manufacturing barriers, pharmacokinetic constraints, and regulatory challenges that must be overcome for successful clinical translation. This multidimensional approach highlights the unique contribution of our review and underscores its relevance for advancing siRNA-based nanomedicine in RA.

In August 2018, Patisiran became the world's first RNA interference (RNAi)–based therapeutic approved in the United States and Europe, marking the advent of a new era in clinical medicine (34). This milestone brought RNAi from preclinical discovery into the clinical spotlight. While gene therapy was traditionally defined as interventions targeting genomic DNA, such as CRISPR-Cas9–based genome editing, RNAi represents an alternative post-transcriptional regulatory strategy, expanding the scope of gene-based therapeutics (35 –37).

RNAi is an evolutionarily conserved mechanism of gene silencing mediated by small RNA molecules. Broadly, RNAi encompasses microRNA (miRNA) and small interfering RNA (siRNA) pathways. miRNAs are endogenous single-stranded RNAs that bind partially complementary sequences on target mRNAs, thereby regulating gene expression by degrading mRNA, reducing its stability, or inhibiting translation (38). In contrast, siRNAs are typically 20–25 nucleotide double-stranded RNAs with perfect complementarity to their target mRNA, resulting in direct cleavage and degradation of the transcript. siRNAs can be either endogenous products of precursor transcripts or exogenously synthesized duplexes.

The canonical RNAi process involves three stages (Figure 1): (i) initiation, in which siRNA precursors are processed by Dicer and exported into the cytoplasm; (ii) effector formation, in which siRNAs are incorporated into the RNA-induced silencing complex (RISC), followed by unwinding and guide-strand selection to target complementary mRNA for degradation by Argonaute 2 in an ATP-dependent manner; and (iii) amplification, in which secondary siRNA synthesis reinforces the silencing cascade, thereby ensuring robust suppression of the target gene.

The therapeutic potential of RNAi in rheumatoid arthritis (RA) has been demonstrated in numerous preclinical studies. Early proof-of-concept experiments employed electroporation-mediated siRNA transfection into joint tissues, successfully silencing TNF-α and alleviating synovial inflammation (39). Since then, a wide range of siRNAs targeting key inflammatory cytokines (e.g., TNF-α, IL-1β, IL-6) or pathogenic pathways (e.g., NF-κB, JAK/STAT) have been designed to suppress inflammation and attenuate joint destruction (40 –43). Interestingly, although IL-6 is a well-established driver of RA pathogenesis, one study found that TNF-α siRNA and IL-1β siRNA produced more pronounced therapeutic effects in collagen-induced arthritis (CIA) mice compared to IL-6 siRNA, underscoring the importance of target selection and disease context (44).

Beyond inflammation, siRNA strategies have also been explored in restoring bone homeostasis and cartilage protection, which represent crucial but often underexplored aspects of RA pathology. Since RA progression involves disruption of the balance between osteoblast-mediated bone formation and osteoclast-mediated bone resorption, targeting molecules in these pathways may provide novel therapeutic avenues (45, 46). Furthermore, cartilage damage—often irreversible in RA—can be mitigated by siRNAs that reduce proteolytic enzyme activity, such as matrix metalloproteinases (MMPs). For example, Tang et al. designed a co-delivery system of MMP-9 siRNA and methotrexate (MTX), which demonstrated synergistic effects in reducing inflammation and reversing cartilage destruction in animal models (18).

Despite the promise of siRNA-based therapies, several challenges hinder their clinical translation. Naked siRNAs are inherently unstable, prone to degradation by serum nucleases, exhibit poor cellular uptake, and may be rapidly cleared via renal filtration. Additional barriers include immune activation, off-target effects, and inefficient delivery to inflamed synovial tissue. To overcome these obstacles, the development of nanocarriers—such as lipid nanoparticles, polymeric nanoparticles, and inorganic nanomaterials—has emerged as a critical strategy. These delivery systems enhance siRNA stability, prolong circulation, improve synovial targeting, and facilitate cytoplasmic release, thereby enabling RNAi to fully realize its therapeutic potential in RA.

Taken together, RNAi represents a transformative approach to RA treatment by specifically silencing pathogenic genes at the mRNA level. While recent advances—including the approval of Patisiran—underscore the translational feasibility of RNAi, the integration of target optimization and nanocarrier-based delivery systems will be essential to advance siRNA therapies from preclinical models to routine clinical application in RA.

Adeno-associated virus (AAV) has long attracted attention as a safe and efficient vector for gene delivery (47, 48). However, its application in the treatment of rheumatoid arthritis (RA) is limited. Given the symmetric, multi-joint involvement characteristic of RA, intra-articular injection is unsuitable for systemic therapy, while systemic administration of AAV suffers from insufficient biodistribution, reducing its therapeutic efficacy. Consequently, AAV may not represent the optimal carrier for siRNA-based therapies in RA. Currently, the most widely used siRNA delivery platforms in clinical studies are modified lipid nanoparticles (LNPs) and N-acetylgalactosamine (GalNAc) conjugates (49). The first FDA- and EMA-approved RNAi therapeutic, Patisiran, employs an LNP delivery system. Although LNPs have demonstrated relatively high transfection efficiency, they still face challenges such as limited tissue penetration, suboptimal cellular uptake, poor ability to cross the blood–brain barrier (BBB), and the requirement for intravenous rather than subcutaneous administration (34). GalNAc conjugates, which bind to the asialoglycoprotein receptor in hepatocytes, markedly enhance siRNA uptake into the liver, but they fail to fundamentally improve siRNA stability or achieve targeted delivery to extrahepatic tissues. In addition, the therapeutic duration of siRNAs remains relatively short, and long-term safety continues to require further validation. These limitations underscore the need for novel delivery systems.

Nanomaterial-based delivery systems (Table 1) provide a promising alternative for siRNA therapy in RA, as they can overcome many drawbacks of traditional carriers, including poor aqueous solubility, low bioavailability, and nonspecific biodistribution (50 –52). Multifunctional nanocarriers can encapsulate sufficient amounts of therapeutic siRNA, prolong circulation half-life, and achieve targeted delivery to inflamed joints. Importantly, such systems can enable stimuli-responsive release based on pathological microenvironmental cues (e.g., acidic pH, elevated reactive oxygen species [ROS], overexpressed proteases) or external triggers such as ultrasound, magnetic fields, or irradiation (53). In addition, theranostic nanocarriers may incorporate imaging agents, allowing real-time monitoring of biodistribution, accumulation, and therapeutic response.

For active targeting, nanocarriers can be functionalized with ligands that bind to overexpressed receptors in inflamed RA joints. For example, hyaluronic acid (HA) targets CD44, folic acid binds folate receptors, mannose ligands recognize macrophage receptors, and RGD peptides bind integrins (Table 2). These modifications significantly enhance cellular uptake in inflammatory synovial macrophages, fibroblast-like synoviocytes (FLS), and neovascular endothelial cells (Figure 2). Such multifunctional nanoplatforms have already shown great potential in cancer, cardiovascular, infectious, and inflammatory diseases, and are now being extended to autoimmune conditions such as RA. By improving delivery efficiency and enabling spatiotemporal control of gene silencing, nano-siRNA carriers provide a new direction for the precision treatment of RA.

The treatment of rheumatoid arthritis (RA) remains a major challenge due to its complex and multifactorial pathogenesis, involving aberrant activation of immune cells, sustained pro-inflammatory cytokine networks, and dysregulated signaling pathways. Although conventional therapies such as methotrexate (MTX), nonsteroidal anti-inflammatory drugs (NSAIDs), and biologics can alleviate symptoms and slow disease progression to some extent, their efficacy is limited and often accompanied by resistance and adverse effects. Thus, the development of safer and more precise therapeutic strategies remains an urgent priority.

RNA interference (RNAi) has emerged as a promising molecular approach for targeted therapy in clinics. Small interfering RNAs (siRNAs), by specifically silencing pathogenic genes, exhibit high precision and selectivity, aligning with the principles of modern precision medicine. Selected siRNA formulations in the clinic (approved/late-stage) were shown in Table 4. Preclinical studies have demonstrated the therapeutic potential of siRNAs in suppressing critical cytokines such as TNF-α, IL-1β, IL-17, and VEGFA, as well as key signaling pathways including NF-κB, JAK/STAT, and MAPK. However, clinical translation of siRNA is hampered by inherent limitations, including instability, rapid nuclease degradation, low cellular uptake, and suboptimal biodistribution. Therefore, the design of efficient and safe delivery systems remains the critical bottleneck for RNAi-based therapeutics.

In recent years, the application of nanocarriers has provided new opportunities to overcome these challenges. Lipid nanoparticles (LNPs), poly(lactic-co-glycolic acid) (PLGA), polyethyleneimine (PEI), and chitosan have been widely investigated and validated as siRNA delivery vehicles for RA therapy. Moreover, emerging platforms such as metal-organic frameworks (MOFs), mesoporous silica nanoparticles, self-assembling peptides, and dendrimers have shown advantages in enhancing siRNA stability, prolonging circulation time, improving site-specific targeting, and achieving stimuli-responsive release. Particularly, intelligent nanoplatforms responsive to pathological features of the RA synovial microenvironment (e.g., acidic pH, elevated ROS, protease overexpression), as well as actively targeted systems modified with ligands such as folic acid, hyaluronic acid, or RGD peptides, have demonstrated significant potential for precise siRNA delivery.

Although current nanocarrier platforms for siRNA delivery—including lipid-based nanoparticles, polymeric micelles, dendrimers, and inorganic nanomaterials — demonstrate promising anti-inflammatory efficacy in RA models, each system exhibits important translational limitations that require critical evaluation. A major challenge is systemic stability: many polymeric and lipid-based vectors undergo opsonization and protein corona formation in circulation, accelerating their clearance by the mononuclear phagocyte system (MPS) and reducing effective half-life (85, 86). Efficient endosomal escape remains another key bottleneck. Cationic polymers such as PEI promote escape through the proton sponge effect but are associated with elevated cytotoxicity, whereas neutral or zwitterionic liposomal systems tend to remain entrapped within endolysosomes, limiting cytosolic siRNA release (87). Off-target biodistribution—particularly hepatic and splenic accumulation—has also been widely reported for both lipid nanoparticles and inorganic nanocarriers due to size- and charge-dependent sequestration (88). Comparatively, liposomes offer excellent biocompatibility but generally suffer from rapid systemic clearance without PEGylation; polymeric nanoparticles (e.g., PLGA, PU, PEG-PLys) provide better structural tunability and sustained release but face hurdles related to protein adsorption, batch consistency, and potential immunogenicity of cationic residues; inorganic nanocarriers such as gold nanoparticles, mesoporous silica, and MOFs offer high loading capacity yet raise concerns regarding long-term biodegradability and tissue retention. Together, these constraints underscore that despite clear advantages in stability, targeted delivery, and intracellular trafficking, the clinical translation of siRNA nanocarriers remains limited by systemic instability, endosomal entrapment, off-target effects, scale-up challenges, and incomplete long-term safety profiles. Future advances—including surface engineering, biomimetic coatings, and multi-responsive architectures—will be essential to push siRNA nanomedicine closer to clinical application in RA.

Several translational challenges remain before these systems can achieve clinical applicability (89). First, regulatory obstacles remain a major barrier. Both the U.S. FDA and EMA require siRNA nanoformulations to meet stringent standards regarding manufacturing reproducibility, nanoparticle size uniformity, sterility, endotoxin levels, and batch-to-batch stability (90). Many nanocarriers used in preclinical studies—such as polymeric micelles, dendrimers, inorganic nanoparticles, and hybrid systems—lack GMP-ready synthesis routes, making scale-up difficult (91). Additionally, regulatory agencies increasingly emphasize comprehensive nanocarrier characterization, including protein corona profiling, complement activation testing, long-term biodegradation behavior, and immunogenicity assessments (92), which remain under-reported in current RA nanomedicine research.

The other one is the formation of a protein corona upon systemic administration. Once exposed to biological fluids, nanoparticle surfaces rapidly adsorb plasma proteins such as albumin, apolipoproteins, and complement components (93, 94). This dynamic corona alters the physicochemical properties of the nanocarriers—affecting particle size, surface charge, and aggregation behavior—and consequently influences biodistribution, cellular uptake, and immune recognition (95, 96). In some cases, the corona may mask targeting ligands or trigger rapid clearance by the mononuclear phagocyte system (MPS), reducing therapeutic efficacy (97). Understanding and controlling protein corona composition through surface engineering (e.g., PEGylation, zwitterionic coatings, or biomimetic membranes) are therefore essential steps toward improving in vivo stability and reproducibility (98, 99).

Third, potential toxicity issues must be systematically evaluated prior to clinical application. Complement activation and the potential for immunotoxicity is a key concern. Positively charged or hydrophobic nanomaterials can activate the complement cascade via the classical or alternative pathways, leading to opsonization, cytokine release, or even hypersensitivity reactions (100, 101). Such immune activation not only compromises biocompatibility but may also exacerbate inflammation in the RA microenvironment (102). Furthermore, many nanocarriers accumulate in the mononuclear phagocyte system (MPS), warranting long-term histopathological studies of liver, spleen, and lymph nodes (103, 104). Without comprehensive toxicological data, regulatory approval for chronic autoimmune diseases remains challenging.

Fourth, dosage considerations are critical for clinical success. Optimizing the pharmacokinetic–pharmacodynamic (PK–PD) relationship is essential because siRNA degradation, rapid renal clearance, and endosomal entrapment can reduce therapeutic efficacy. Pharmacokinetic properties—such as circulation half-life, biodistribution, and metabolic clearance—directly influence pharmacodynamic outcomes including target gene silencing efficiency and duration of anti-inflammatory effects. For instance, nanocarriers with prolonged systemic retention and controlled siRNA release profiles can maintain sustained therapeutic concentrations at inflamed joints while minimizing off-target exposure (10, 105). Conversely, rapid clearance or premature siRNA degradation may result in suboptimal pharmacodynamic responses despite high in vitro delivery efficiency. Such data are largely absent from current RA-related siRNA studies.

Fifth, although clinical development of siRNA therapeutics has accelerated, clinical trials directly targeting RA remain limited. Several siRNA drugs—such as patisiran, givosiran, inclisiran, and vutrisiran—have gained FDA approval for other diseases, (Table 4)demonstrating the clinical feasibility of RNAi therapeutics (34, 106 –108). These first-in-class examples provide regulatory precedents for formulation design, lipid nanoparticle (LNP) or GalNAc conjugate platforms, and validated safety profiles (109, 110). However, no siRNA nanomedicine has yet progressed to advanced clinical trials for RA. Existing early-stage studies primarily address inflammatory diseases or fibrotic disorders rather than autoimmune arthritis, highlighting an unmet translational gap.

Sixth, the substantial heterogeneity of the synovial microenvironment across different stages of RA is another important underappreciated barrier to clinical translation. Early RA is characterized by highly vascularized synovial tissue, abundant infiltrating immune cells, and a relatively loose extracellular matrix (ECM) architecture, which may facilitate deeper nanocarrier penetration and more efficient siRNA uptake by macrophages and fibroblast-like synoviocytes (111). In contrast, late-stage RA presents a markedly altered microenvironment: chronic inflammation leads to fibrotic pannus formation, increased ECM density, hypoxic niches, reduced vascular perfusion, and the emergence of distinct macrophage and fibroblast phenotypes (112, 113). These structural and cellular changes can impede nanoparticle diffusion, reduce endocytic activity, alter intracellular trafficking, and modify the pharmacodynamic response to siRNA (114). Moreover, cytokine gradients and receptor expression profiles (e.g., CD44, folate receptor β, integrins) evolve dynamically during disease progression (115, 116), potentially affecting the targeting efficiency of ligand-modified nanocarriers. Recognizing these temporal variations underscores the need for stage-adapted or microenvironment-responsive delivery systems, as a single nanocarrier design may not achieve uniform efficacy throughout the RA disease course.

Another major factor influencing the clinical translation of siRNA nanotherapeutics is the significant patient-to-patient variability in rheumatoid arthritis, which extends beyond disease stage and affects multiple biological determinants of therapeutic response. RA patients exhibit substantial differences in synovial cellular composition—such as variable proportions of macrophage subsets, fibroblast phenotypes, and infiltrating lymphocytes—which can alter nanoparticle uptake profiles and intracellular trafficking pathways (117). Genetic polymorphisms affecting cytokine production, RNAi pathway machinery, or endocytic receptors (e.g., CD44, folate receptor β, scavenger receptors) may further contribute to heterogeneity in target knockdown efficiency and ligand-mediated targeting specificity (118). Circulating and synovial cytokine signatures also differ markedly between individuals, shaping inflammatory gradients and modifying the physicochemical landscape encountered by nanocarriers (119). Moreover, inter-individual variability in pharmacokinetics—such as renal clearance rate, hepatic metabolism, complement activation tendency, and mononuclear phagocyte system (MPS) activity—may result in inconsistent biodistribution and exposure levels of siRNA formulations (120). Collectively, these factors underscore the need for personalized or stratified nanocarrier design, potentially incorporating biomarker-guided patient selection, adaptable targeting ligands, and microenvironment-responsive delivery strategies to achieve consistent therapeutic outcomes across heterogeneous RA populations.

Overall, siRNA-based nanomedicine represents a promising therapeutic paradigm for rheumatoid arthritis; however, its clinical translation remains constrained by several important challenges. Although current nanocarrier platforms demonstrate strong gene-silencing efficiency and encouraging anti-inflammatory outcomes in preclinical models, their behavior in humans is far less predictable. Key obstacles—including systemic stability, protein corona formation, endosomal escape efficiency, batch-to-batch manufacturing consistency, and long-term biosafety—must be rigorously addressed before clinical application can be realized. In addition, the heterogeneous nature of RA and its fluctuating inflammatory microenvironment imply that a single siRNA target or delivery strategy may not be universally effective across patient populations. Regulatory considerations also pose significant barriers, as siRNA nanomedicines must meet strict requirements for GMP production, quality control, sterility, pharmacokinetics, immunogenicity, and degradation profiling.