1. Introduction

Pancreatic cancer broadly refers to all malignant tumors that originate from the epithelial tissue of pancreas, with the most common subtype being pancreatic ductal adenocarcinoma (PDAC). PDAC accounts for 80%-90% of all pancreatic cancers^[1], and holds infamous distinction of being labeled the “king of cancers”. The deep anatomic location of pancreas contributes to subtle and non-specific early symptoms, inadequate surgical resection rates, and limited efficacy of chemotherapy, collectively posing significant challenges for developing efficacious therapeutic agents for pancreatic cancer^[2]. RNA therapy is regarded as a powerful therapeutic method with its myriad biological functions, which has contributed to advances in treating various diseases such as cancers, viral infections, and inflammatory disorders in the last few years. Various therapeutic RNAs, such as small interfering RNA (siRNA), antisense oligonucleotide (ASO), messenger RNA (mRNA), have been used in gene therapy for several diseases. While these RNA-based drugs have led to successful therapeutic outcomes, there is still room for improvement in several aspects such as blood circulation time, stability, and RNA loading capacity. For example, RNA-based therapeutics are highly susceptible to degradation by serum nucleases in bloodstream^[3]. Intravenously administered nanoparticles or delivery vehicles are readily taken up and cleared by the reticuloendothelial system, particularly in liver and spleen^[4]. This significantly reduces the effective drug dose that reaches pancreas; After cellular uptake, RNA drugs are typically trapped within endosomes and ultimately transported to lysosomes where they are degraded by enzymes^[5]. To exert their function, RNA-based drugs (especially siRNA/mRNA) must effectively escape from the endosomes with the help of a carrier and be released into the cytoplasm.

This article aims to systematically delineate the application of RNA-based therapeutics for pancreatic cancer, with a focus on three pivotal aspects: The challenges of pancreatic cancer treatment, existing strategies for pancreatic-targeted delivery strategies, and the latest advancements in the utilization of RNA-based therapeutics for the treatment of pancreatic cancer (Scheme 1).

Display Full Size

Download

Scheme 1. Overview of delivering RNA-based therapeutics for pancreatic cancer treatment. Created in BioRender. Chang, L. (2026) https://BioRender.com/r5om53y. LNP: lipid nanoparticle; ASO: antisense oligonucleotide; siRNA: small interfering RNA; mRNA: messenger RNA.

2. Challenges of Pancreatic Cancer Treatment

This intricate convergence of anatomical constraints, pathophysiological aberrations, and immunosuppressive mechanisms orchestrates a robust defense that renders pancreatic tumors refractory to standard therapeutic modalities. The pancreas, with its unique lobulated architecture and stringent blood-pancreas barrier, presents a formidable fortress that limits drug accessibility. Within this tumor microenvironment (TME), tumor cells thrive amidst a densely desmoplastic stroma orchestrated by tumor-associated fibroblasts, while aberrant vascularization and a profoundly immunosuppressive microenvironment further impair therapeutic efficacy. This chapter systematically dissects these interconnected challenges, including the physical barricades of the blood-pancreas barrier and extracellular matrix, as well as the cellular conspirators of cancer-associated fibroblast (CAF).

2.1 Blood pancreas barrier (BPB)

The treatment of pancreatic cancer, particularly PDAC, faces a formidable obstacle in form of BPB. This highly selective physiological barrier severely restricts the penetration of both small and large molecules into pancreas, drastically limiting distribution and efficacy of chemotherapeutic drugs^[6]. The BPB is a complex, multi-layered structure that includes pancreatic capillary endothelial cells with tight junctions, a continuous basement membrane, and a unique periacinal fibroblastic sheath, which is a prominent structure that physically separates capillaries from acinar cells. Furthermore, the acinar cell basement membrane and the tight junctions between acinar cells themselves add additional layers of defense. Consequently, this intricate barrier system presents a major challenge in developing effective therapeutics.

2.2 Pancreatic TME

TME is a complex ecosystem composed of both cellular and non-cellular components that surround tumor tissue. Its complex tissue structure creates a physiological barrier that prevents targeted drugs from reaching tumor sites. The following section will introduce the key components of the pancreatic TME.

2.2.1 Pancreatic tumor extracellular matrix (ECM)

ECM, one of the sources of physiologic resistance of cancer cells to treatment, is regarded as a serious trouble for anti-cancer approaches such as gene and immune therapies. ECM stiffness is obstructive for effective uptake or delivery of drugs to the intratumoral area (Figure 1). High deposition of collagen proteins would do this possibly through binding to proteoglycans and further stabilizing these components of ECM^[7]. Increased cancer stem cell (CSC) stemness and expansion of the stem cell niche in TME, mediated by ECM stiffness, hamper drug penetration to this niche^[8]. It is hypothesized that ECM stiffness is associated with transformation of cancer cells toward CSCs that can survive under hypoxic conditions of the TME. These stem-like cells are also more resistant to cytotoxic drugs, and they have the potential to migrate and invade via surrounding tissues^[9]. In addition, ECM stiffness stimulates hypoxic conditions within the TME. Hypoxic microenvironment triggers abnormal tumor vascular functions including increased permeability and disrupted vascular structure, which raise vascular rupture and leakage risks, and facilitate tumor cell intravasation and distant metastasis. Meanwhile, hypoxia hinders normal vascular maturation and results in immature neovessels, ultimately impairing blood perfusion and drug delivery efficiency in tumor tissues^[10].

Display Full Size

Download

Figure 1. Tumor extracellular matrix reduces therapeutic efficiency in pancreatic tumors. Created in BioRender. Chang, L. (2026) https://BioRender.com/rsxzi9y. CAFs: cancer-associated fibroblasts.

2.2.2 CAFs

Fibroblasts represent the most abundant cell type in the TME and are largely responsible for the deposition of the extracellular matrix, driving characteristic physiological properties of PDAC. CAFs display substantial cellular heterogeneity that critically complicates tumor progression and metastatic dissemination. Multiple cellular origins, including tissue-resident fibroblasts, mesenchymal stem cells^[11], and endothelial cells^[12], can transdifferentiate into CAFs. CAFs characteristically upregulate α-smooth muscle actin, platelet-derived growth factor receptor α, and fibroblast activation protein, thereby directly fostering tumor cell invasion and proliferation^[13]. Furthermore, CAFs comprise multiple distinct subtypes. Among them, myofibroblastic CAFs drive ECM remodeling and mechanical microenvironment modulation; inflammatory CAFs secrete immunosuppressive cytokines; and antigen-presenting CAFs (apCAFs) induce immune tolerance^[14]. Notably, apCAFs are broadly distributed and can actively promote tumor progression in PDAC^[15].

CAFs have the capacity to secrete a variety of ECM-related proteins, including collagen and hyaluronic acid, which form a physical and metabolic barrier, thereby reducing the clinical effect of chemotherapy and immunotherapy on PDAC, and promoting tumor growth and invasion^[16]. CAFs have also been observed to secrete a variety of cytokines, which have been linked to the promotion of tumor invasion, metastasis and chemoresistance^[17]. Furthermore, these cells have been shown to modify the phenotype of tumor tissue, ultimately resulting in the development of proliferative fibrosis within the connective tissue of the pancreas. Through their role in negative immune regulation, they can compromise the efficacy of immunotherapy. Strategically designed therapeutic interventions targeting the suppression of CAFs genesis or the disruption of CAFs functionality may effectively reverse their negative immunomodulatory effects within the tumor microenvironment, thereby potentiating the efficacy of immunotherapeutic approaches.

2.2.3 Hypervascularization of PDAC

Insufficient vascularization within pancreatic tumor tissue hinders drug delivery through the circulatory system. The presence of a dense extracellular matrix and elevated interstitial pressure has been demonstrated to result in the collapse of newly formed blood vessels. The glycolytic metabolism driven by KRAS mutations consumes local oxygen, further exacerbating vascular dysfunction^[18].

2.2.4 Immunosuppressive properties of pancreatic tumors

Pancreatic cancer exhibits a profoundly immunosuppressive microenvironment characterized by a paucity of functional CD8⁺ cytotoxic T lymphocytes, restricted to the tumor periphery, scarce and dysfunctional dendritic cells, and extensive infiltration of Tregs, tumor-associated macrophages (TAMs), and myeloid-derived suppressor cells (MDSCs). Elevated transforming growth factor-β, interleukin (IL)-10, and vascular endothelial growth factor further consolidate this immunoinhibitory milieu, collectively rendering immunotherapy largely ineffective^[19]. These multifaceted challenges necessitate innovative delivery platforms capable of simultaneous barrier penetration, stromal remodeling and immune activation, a promise that emerging nanotechnology seeks to fulfill.

3. Pancreatic Targeted Delivery Strategy

Exposed RNA nucleic acids (siRNA, mRNA and ASO) are easily and rapidly broken down by nucleases in plasma and tissues in the circulatory system, resulting in a loss of drug efficacy. At the same time, due to their electronegative nature, nucleic acid molecules find it difficult to enter target cells autonomously. Furthermore, the naked molecules accumulate easily in the liver, causing liver injury. Therefore, nucleic acid drug targeting of pancreatic organs must be achieved using a drug delivery system (Figure 2).

Display Full Size

Download

Figure 2. Delivery particle for pancreatic cancer therapy. Created in BioRender. Chang, L. (2026) https://BioRender.com/ue749o6. AAV: adeno-associated virus; LNP: lipid nanoparticle.

3.1 Exosomes delivery system

Exosomes are membrane-bound vesicles that are released into the extracellular matrix when intracellular multivesicular bodies fuse with the cell membrane. They can transport a variety of substances, including proteins, lipids, DNA and RNA. Using extracellular vesicles (EVs) as RNA-based drug delivery carriers targeted to the pancreas has the following advantages: 1. The surface of EVs can be equipped with specific membrane proteins, enabling them to actively target specific cells or tissues in the pancreas and pancreatic tumors^[20]; 2. The extracted exosomes have a natural homing effect towards the pancreas^[21]; 3. Derived from natural cells, exosomes can reduce immune rejection reactions and improve in vivo stability^[22]; 4. They possess a transmembrane transport capacity and can cross the BPB.

The following table illustrates several methods for the realization of exosome delivery to the pancreas (Table 1). One method involves chemical modification of oxaliplatin prodrugs to the surface of exosome particles, thus enhancing their targeting ability towards pancreatic tumors^[23]. Another method uses the CD47 or CD81 protein on the surface of exosome particles to inhibit phagocytosis by binding to SIRPα on monocytes and macrophages^[24]. A further method involves the genetic engineering of exosome-producing source cells to secrete exosome particles that possess CD133 targeting properties, thus enabling the targeting of pancreatic tumors^[27]. Finally, the co-displaying of Arg-Gly-Asp (a peptide capable of binding to αvβ3 integrin on the surface of tumor cells) on exosome particles enhances pancreatic tumor tropism^[28].

The following must be considered when transitioning exosomes into clinical practice: 1. The efficacy and safety of the cell line used for exosome production; 2. The stability of the exosomes following prolonged freeze-thaw cycles; 3. The potential in vivo toxicity associated with long-term administration^[29]. To overcome these challenges, parental cells can be genetically engineered to express pancreatic-homing ligands, standardized large-scale production protocols can be developed, and exosome-mimetic nanovesicles can be constructed to improve targeting and controllability.

3.2 Pancreatic targeted polymeric nanodelivery systems

In recent years, there has been a notable increase in the application of nanotechnology in the pharmaceutical field. The application of nanotechnology involves the utilization of nanomaterials for the precise delivery of nucleic acid drugs to pancreatic tissue. Nano polymer offers unparalleled tunability in size, surface chemistry, and stimuli-responsive behavior, enabling rational design to effectively overcome lobular heterogeneity, dense desmoplastic stroma, and the stringent blood-pancreas barrier, and even navigate the heterogeneous TME^[30]. Their facile adaptability for sophisticated surface engineering and stimuli-responsive design renders them particularly promising candidates to surmount the biological barriers that have historically constrained pancreatic targeting efficiency.

The nano polymer is characterized by its minute size (less than 200 nm) and physicochemical properties that render it particularly well-suited for nanomedical applications^[31]. These properties have been shown to enhance membrane permeability and the retention effect. The small size and alterable surface specificity of these nanoparticles enable them to deliver therapeutic drugs to targeted cancer cells in the biological system via cellular uptake. This uptake can occur via various mechanisms, including receptor-mediated uptake and endocytosis, known as active and passive transport mechanisms^[32].

The methods employed to achieve targeted delivery to pancreatic tumors can be categorized into two primary types: passive targeting and active targeting. Passive targeting enhances accumulation within pancreatic tumors by modifying nanoparticle size and composition. Active targeting achieves specificity by conjugating nanoparticles with specific ligands that bind to receptors uniquely expressed on pancreatic tumors’ cells. The subsequent table provides a selection of nanoparticle delivery systems targeted to the pancreas (Table 2).

The disadvantages of polymeric nanocarriers are also pressing issues requiring resolution. For instance, certain polymeric components may elicit innate immune responses, accelerate clearance kinetics and introduce risks of adverse reactions. Furthermore, the long-term fate of non-biodegradable or slowly degrading nanoparticles within pancreatic tissue remains poorly characterized, with potential risks of lysosomal dysfunction or chronic inflammation^[55]. To overcome these challenges, naturally biodegradable polymers should be selected, and pancreatic-targeting moieties should be conjugated to enhance specific delivery to pancreatic tissue.

3.3 Pancreatic targeted lipid nanoparticle (LNP) delivery system

LNPs are advanced delivery systems designed to encapsulate and transport nucleic acids, utilizing a self-assembly process driven by microfluidic mixing of lipid-containing organic phases and RNA-containing aqueous phases (Figure 3). Their key advantages include high encapsulation efficiency, strong protection of RNA therapeutics, with their clinical utility well exemplified by COVID-19 mRNA vaccines. LNPs also hold promise for gene therapies and cancer immunotherapies. The pathophysiological features of PDAC, including dense desmoplasia, acidic hypoxia, CAF dominance, and profound immunosuppression, present formidable yet addressable challenges for therapeutic delivery. LNPs, with their precisely tunable physicochemical properties, pH-responsive activation mechanisms, facile surface functionalization for CAF targeting, and intrinsic immunostimulatory capacity, represent a technologically mature platform, ideally suited to exploit these tumor-specific vulnerabilities.

Display Full Size

Download

Figure 3. Flow chart of LNP encapsulated nucleic acid drugs. Formulation: After sequence design, the synthesized bare nucleotides are dissolved in the aqueous phase, while phospholipid molecules are dissolved in the organic phase. The two phases self-assemble into drug-loaded LNPs through a microfluidic device. Characterization: Examine size; PDI; stability; encapsulation efficiency; Zete potential of the formulated LNP. Created in BioRender. Chang, L. (2026) https://BioRender.com/3pu1b12. LNP: lipid nanoparticle; PDI: polydispersity index; PEG: polyethylene glycol; mRNA: messenger RNA; ELS: electrophoretic light scattering.

However, a major limitation of conventional LNPs is their strong hepatic tropism after intravenous injection, which restricts delivery to non-liver tissues such as the pancreas^[56]. To improve pancreatic targeting (Table 3), strategies include optimizing lipid formulations (e.g., using novel cationic lipids like P6CIT or adjusting cholesterol content^[57]), modifying surface chemistry with targeting ligands (e.g., tLyp-1 peptide for tumor homing or vitamins for enhanced pancreatic uptake^[60]), and controlling particle size. For instance, smaller (~30 nm) LNPs improve general pancreatic distribution, while larger ones (~160 nm) favor islet retention^[61].

Although these methods have achieved pancreatic targeting, the targeting efficiency remains low. In particular, the delivery efficiency of intravenously administered pancreatic-targeting systems based on particle size modulation is less than 10%. How to achieve precise and efficient pancreatic targeting remains a technical barrier that needs to be overcome at this stage. Additionally, for strategies involving ligand-modified nanoparticles for pancreatic targeting, identifying suitable ligands has also become a significant challenge. The intricate multicomponent architecture of LNPs further increases the difficulty of formulation screening and optimization. Notably, several pioneering studies have demonstrated innovative strategies to achieve pancreatic targeting using LNPs. Ivan Isaac et al. reported a breakthrough approach by incorporating vitamins as a fifth component into the LNP formulation, leveraging vitamin D receptor-mediated endogenous targeting mechanisms to enable specific pancreatic delivery^[59]. In parallel, Jillian R. Melamed et al. achieved pancreatic mRNA expression through intraperitoneal injection, circumventing the hepatic first-pass effect^[58]. Jiaqi Lei et al. optimized pancreatic delivery by modulating LNP particle size, demonstrating that size-tuned LNPs administered via intraperitoneal route could efficiently penetrate pancreatic tissue^[63]. Qiu et al. optimized an ionizable lipid (4A2-B8-PH) that achieves pancreatic mRNA expression^[62]. These multifaceted approaches encompassing vitamin receptor targeting, alternative administration routes, and physicochemical optimization converge to enable clinically viable pancreatic mRNA delivery.

3.4 Viral vector delivery system

The utilization of viral vectors for nucleic acid drug delivery has several advantages. The primary advantage of viral vectors lies in their ability to target specific organs with a high degree of precision^[64]. There are several methods to improve pancreatic targeting (Table 4). For instance, the serotype of a given recombinant adeno-associated virus (AAV) can be modified through the alteration of the cap gene within the packaging plasmid, thereby altering its tissue tropism or susceptible cell types^[66]. As demonstrated in preceding studies, AAV6, AAV8, AAV9 and AAV pan exhibit a high degree of infectivity in the pancreas. Among these, AAV8 has the greatest prevalence, while AAV6 demonstrates optimal infectivity in the pancreatic duct^[72]. Viral vectors possess the capacity for effective penetration of biological tissue, a property advantageous for their ability to traverse the BPB.

Although viral vectors have achieved efficient pancreatic-targeted delivery, their relatively slow expression kinetics make them more suitable for long-term expression scenarios, failing to meet the requirements of mRNA for short-term, high-dose expression. Moreover, viral vectors face challenges in packaging long-chain mRNA or chemically modified mRNA molecules^[73]. They also may trigger pre-existing immune responses (e.g., adenoviral vectors), leading to compromised delivery efficiency or adverse side effects.

3.5 Antibody delivery system

Antibodies have garnered significant recognition in the treatment of various diseases, particularly cancer, owing to their exceptional specificity, high affinity for target cell antigens, and prolonged circulation in the body.

With the further advancement of antibody-mediated RNA delivery technology, increasing researchers are turning their attention to the development of antibody-oligonucleotide conjugate (AOC) technology. AOC demonstrates enhanced pharmacokinetic properties and more precise biodistribution compared to traditional oligonucleotide therapy. It can be conjugated to various types of oligonucleotides, including siRNA and ASO. This flexibility allows researchers to design different AOC drugs according to different disease types^[74]. Currently, clinical trials are underway for AOC drugs, showcasing their potential in therapeutic applications.

The latest research has discovered TMAB3, a tumor-targeting monoclonal antibody which exhibits robust cell penetration and the capacity to bind RNA. TMAB3 can form stable, non-covalent antibody/RNA complexes of a defined size, enabling highly specific and functional RNA delivery within tumors. Single-cell RNA sequencing and flow cytometry revealed that TMAB3/3p-hpRNA treatment evoked a robust antitumor immune response characterized by RIG-I activation and increased infiltration and activation of cytotoxic T cells. These findings establish that the TMAB3/RNA complex can deliver RNA payloads specifically to hard-to-treat tumor cells, achieving anti-tumor effects and providing an antibody-based platform to advance RNA therapy research in cancer patients^[75].

4. Treatment Strategies for Pancreatic Cancer

4.1 Conventional treatment methods for pancreatic cancer

Current standard-of-care for pancreatic cancer encompasses cytotoxic chemotherapy, immunotherapy, and supportive interventions. Gemcitabine, fluorouracil, nab-paclitaxel, and paclitaxel constitute first-line antiproliferative agents. Oxaliplatin and irinotecan are reserved for advanced disease in combination protocols. Checkpoint inhibitors such as nivolumab and pembrolizumab reactivate antitumor immunity, though durable responses remain subtype specific. Pancreatic enzyme replacement therapy addresses exocrine insufficiency and improves quality of life. Oral capecitabine-based regimens offer palliative options for late-stage disease, yet variable bioavailability necessitates careful patient selection^[76].

Although conventional therapies have achieved modest efficacy in pancreatic cancer treatment, chemotherapy is constrained by off-target toxicity, stromal barrier exclusion, and acquired drug resistance^[76]. Immune checkpoint inhibitors demonstrate limited clinical benefit due to the immune-desert phenotype and immunosuppressive microenvironment^[77]. Oral formulations suffer from variable bioavailability and heterogeneous patient responses. RNA therapeutics, as an emerging technology platform, offer substantial promise by enabling sequence-specific gene regulation, de novo protein synthesis, and rapid therapeutic adaptation, thereby presenting a transformative paradigm to overcome the limitations that have historically plagued pancreatic cancer management.

4.2 RNA-based therapeutic for the treatment of pancreatic cancer

4.2.1 mRNA therapeutics

mRNA therapy represents a therapeutic strategy wherein in vitro-transcribed mRNA is introduced into host cells to direct the transient production of therapeutic proteins^[78]. Key structural components of synthetic mRNA ensure its stability and efficient translation: the 5’ cap protects against exonuclease degradation and initiates translation, while the optimized 5’ and 3’ UTRs enhance ribosome binding and prolong mRNA half-life^[79]. A long poly-A tail further stabilizes the molecule and strengthens translational persistence^[80].

The major advantage of mRNA drugs rests with their ability to utilize the host’s cellular machinery to produce complex or secreted proteins without risking genomic integration, enabling rapid and transient protein expression. This makes them highly versatile for treating various conditions. For example, cancer occurs when genes in normal cells mutate, leading to abnormal cell proliferation. Some unique proteins are produced on the surface of cancer cells, known as ‘tumor antigens’^[81]. The design of mRNA vaccines is based on these tumor antigens. By delivering mRNA that encodes tumor antigens to human cells, the cells synthesize the antigens according to the mRNA’s instructions^[82]. Once these antigens have been synthesized, immune cells recognize them and activate the immune system, triggering a series of immune reactions. T cells are activated and can recognize and attack cancer cells specifically^[83], while B cells produce antibodies to further enhance the immune response. Additionally, the mRNA vaccine activates the immune system’s memory function, enabling it to attack cancer cells faster and more effectively if they are encountered again in the future, thereby preventing recurrence.

mRNA tumor vaccines fall into two categories. One category is individualized cancer vaccine, which recognizes specific antigens according to the tumor tissues of patients (Figure 4). mRNA technology is used to produce specific cancer antigens in vivo and activate the T-cell immune response to kill tumor cells. The other is the universal cancer vaccine, which uses mRNA technology to encode a broad spectrum of tumor-associated antigens and activate the immune system to kill tumor tissue^[84]. The subsequent table will present recent advances in clinical (Table 5) and preclinical (Table 6) research on mRNA technologies for pancreatic cancer treatment.

Display Full Size

Download

Figure 4. The workflow of PCV. (a) Collect tumor tissue samples and peripheral blood samples from patients; (b) Compare the DNA of tumors and normal cells to identify tumor specific non synonymous mutations; (c) Determine the patient’s HLA subtype through second-generation sequencing to ensure that the subsequently screened antigens can be recognized by the patient’s immune system; (d) Design mRNA sequences that can translate specific antigens through a certain codon encoding method; (e) Preparation of mRNA vaccines; (f) Activate the body’s immune response. Created in BioRender. Chang, L. (2026) https://BioRender.com/ygrqe4f. PCV: personalized cancer vaccine; mRNA: messenger RNA; HLA: human leukocyte antigen.

Protein replacement therapy represents another widely employed strategy for pancreatic cancer treatment, among which cytokine supplementation constitutes a representative approach. Qiu et al. reported the treatment of pancreatic cancer via intraperitoneal injection of 4A2-B8-PH LNPs encapsulating IL-12 mRNA^[62]. Once released into the TME, IL-12 binds to the its receptor on the surface of T cells and NK cells, activating downstream signaling pathways such as the JAK-STAT pathway. This activation promotes the proliferation, differentiation, and cytotoxicity of T cells (particularly CD8⁺ cytotoxic T lymphocytes) and NK cells. Anti-PD-1 antibodies are widely used clinically as immune checkpoint inhibitors. mRNA encoding PD-1 antibodies enables transient expression of PD-1 antibodies within host cells, blocking the interaction between PD-1 and PD-L1 and preventing T cell exhaustion, thereby achieving therapeutic efficacy in pancreatic cancer^[89].

Furthermore, mRNA technology can be employed to weaken the fibrotic barrier on the surface of tumor tissue. Using a technique involving a mannose-modified mRNA-LNP, M2 macrophages can be reprogrammed in vivo into FAP-CAR-M. Following adoptive transfer, FAP-CAR-Ms infiltrate the desmoplastic stroma and mediate selective depletion of CAFs through antibody-dependent cellular phagocytosis, thereby alleviating physical compression of tumor vasculature and enhancing drug penetration. Concurrently, CAR-M polarization shifts toward pro-inflammatory M1 phenotypes, reversing immunosuppression and fostering T cell infiltration, thereby enhancing the efficacy of immunotherapy^[88].

4.2.2 siRNA therapeutic

Structural characteristics of siRNA drugs: siRNA is composed of two complementary strands, the guide strand (antisense strand) and the passenger strand (sense strand). Its 3’ end has two hydroxylated nucleotide overhangs (such as 3’ Oh dinucleotide), enhancing stability and RNA induced silencing complex (RISC) recognition efficiency^[90].

Mechanism of action of siRNA drugs: siRNA is loaded into RISC, in which argonaute2 protein unwinds the siRNA, degrades the passenger strand, and retains guide strand as targeting guide^[91]. siRNA can not only guide RISC to cleave homologous single-stranded mRNA, but also serve as a primer to bind to target RNA and synthesize more dsRNA under the action of RNA-dependent RNA polymerase (RdRP)^[92]. The newly synthesized dsRNA is then cleaved by Dicer, generating many secondary siRNA, thereby amplifying the effect of RNAi.

In the process of growth and metastasis of pancreatic cancer, there will be many genes with abnormal expression and upregulation. siRNA drugs can selectively silence these genes with high abnormal expression to play an anti-tumor role. The primary targets of these drugs concentrate on three types of pathways: the tumor proliferation and metastasis pathway, the tumor resistance pathway, and the immune pathway. For example, the following tables will introduce some related targets for siRNA therapeutic (Table 7) and clinical trials for pancreatic treatment (Table 8).

4.2.3 ASO therapeutic

ASO therapy is a small-nucleic acid drug technology that regulates gene expression by specifically binding to target mRNA. The core mechanism of this process involves the formation of complementary base pairing between chemically modified oligonucleotide sequences and target RNA. A DNA-RNA hybrid duplex can be formed through complementary base pairing with specific sequences on the target mRNA^[112]. The RNase H1 enzyme specifically cleaves the RNA part of the hybrid strand, resulting in the degradation of the target mRNA. The cleaved and degraded mRNA cannot be translated into protein, thereby downregulating the expressing of the targeted RNA^[113].

ASO drugs, like siRNA drugs, can also silence the expression of targeted genes. However, the difference lies in that after siRNA degrades target mRNA through the RISC complex, RISC can repeatedly bind to new targets, extending the duration of drug efficacy^[114]. In addition, the released sense strand of siRNA can serve as a primer to amplify dsRNA under the action of RdRP. After degradation by Dicer, more siRNA is generated, achieving a cascade amplification effect and maintaining the effect for 3-6 months^[115]. ASO mainly degrades mRNA through RNase H1 or interferes with spatial steric hindrance, lacking a similar cyclic amplification mechanism; its effect is a single event with a half-life of only 2-4 weeks, and the target level tends to rebound after drug withdrawal^[116]. ASO drugs targeting pancreatic tumors currently focus on three types of pathways: tumor growth and metastasis pathway, tumor drug resistance pathway, and immune evasion pathway. The following table would show some related research on ASO therapy (Table 9 and Table 10).

5. Prospects

Thanks to continuous breakthroughs in delivery technology, sequence screening and nucleic acid structure modification techniques, the field of RNA-based drugs is expanding beyond traditional viral vaccines to include areas such as cancer and metabolic diseases^[132]. This article provides a comprehensive overview of the latest research on the use of nucleic acid drugs in treating pancreatic cancer. However, several issues still need to be addressed before pancreatic-targeted nucleic acid drugs are clinically translated. The production cost of exosomes is high, and their large-scale production remains difficult to achieve^[133]. Nano polymers are not stable enough, and their long-term safety in the human body is unknown^[134]. Although pancreatic-targeted liposomes improve drug targeting to the pancreas, production cost also increases^[135]. Viral vectors can trigger human immune storms and inflammatory reactions^[136]. Polymeric nanocarriers and LNPs enable precise, scalable manufacturing with well-defined quality attributes compared to exosomes and viral vectors. Polymeric systems offer exceptional chemical tunability, as monomer composition, molecular weight, and crosslink density can be precisely manipulated, whereas LNP lipid formulations are constrained by phase behavior and stability requirements. Nevertheless, LNPs demonstrate superior endosomal escape efficiency through the proton sponge mechanism and benefit from established clinical infrastructure, providing regulatory precedent and manufacturing scalability that polymer platforms currently lack.

Overcoming the BPB remains a critical challenge for effective RNA therapeutics in pancreatic diseases. The BPB comprises capillary endothelial cells, basement membranes, and pancreatic stellate cells that restrict macromolecular entry^[137]. Nanoparticles can exploit small-size effects, typically below 200 nm, to penetrate intercellular tight junctions and breach this barrier. Surface functionalization with ligands targeting highly expressed pancreatic receptors (such as vitamin D), can further enhance nanoparticle tropism. In tumor-bearing contexts, pancreatic cancer cell-associated antibodies or acid-sensitive formulations can improve enrichment within the TME. Alternatively, intraperitoneal administration offers a promising strategy to bypass the BPB entirely and achieve direct pancreatic delivery^[138].

Current research into nucleic acid drug therapy mostly focuses on single target. Multi-drug combination therapy strategies require further exploration. Due to the limitations of current delivery technologies, drug accumulation in other organs can easily lead to adverse reactions unrelated to therapeutic purpose. Future development should focus on enhancing drug targeting to the pancreas and exploring multi-target combination therapy.

6. Conclusions

siRNA, mRNA, and AOC have emerged as promising therapeutic tools for PDAC. However, the unique biological barriers of PDAC and inherent limitations of delivery systems have severely constrained their clinical applications. The distinctive biological barriers of PDAC present multi-layered challenges for nucleic acid delivery. The blood-pancreas barrier, characterized by tight junctions and selective permeability, restricts the delivery of nucleic acid drugs to pancreatic tumor sites. The immunosuppressive TME, featuring M2-type TAMs and MDSCs, further compromises the efficacy of immunomodulation-dependent nucleic acid therapeutics. Additionally, the dense desmoplastic stroma creates a physical barrier that impedes therapeutic agent penetration into the tumor parenchyma. Efficient delivery systems capable of overcoming PDAC barriers are essential for realizing the therapeutic potential of nucleic acid-based treatments in pancreatic cancer.

Existing nucleic acid delivery systems demonstrate varying capabilities in addressing PDAC-specific delivery challenges, each with distinct advantages and inherent limitations. Cell-derived exosomes, as natural intercellular communication vehicles, possess inherent biocompatibility and the ability to traverse biological barriers. However, the low yield of native exosomes and difficulties in large-scale production have hindered their clinical translation. Polymeric carriers and LNPs offer tunable physicochemical properties, and their tumor-targeting capability can be enhanced through the modification with ligands (such as peptides or antibodies) targeting PDAC-specific antigens. These carriers can protect nucleic acids from nuclease degradation in the circulation and TME. Yet their penetration into the dense PDAC stroma remains limited, and non-specific accumulation may lead to off-target toxicity. Viral vectors, despite their high transfection efficiency, face significant immunogenicity concerns in the immunosuppressive TME of PDAC, potentially triggering adverse immune responses and compromising long-term therapeutic efficacy.

To improve delivery efficiency, future strategies must integrate pathological insights of PDAC with targeted delivery system optimization. First, enhancing stromal penetration is critical. For instance, modifying delivery vehicles with enzymes such as hyaluronidase can degrade extracellular matrix components and reduce stromal density to facilitate carrier penetration into the tumor parenchyma. Second, precise tumor targeting can be achieved by functionalizing carriers with ligands that recognize PDAC-specific molecules. Alternatively, designing pH-sensitive or enzyme-responsive polymeric/LNPs enables specific triggering of nucleic acid release in the acidic TME or upon encountering stroma-specific enzymes, thereby improving intracellular drug bioavailability.

In conclusion, nucleic acid drug delivery for PDAC treatment faces unique challenges arising from tumor biological barriers and pathological complexity. Existing delivery systems have made progress in addressing specific barriers. However, further optimization is needed to enhance stromal penetration, tumor targeting, and intracellular bioavailability. Future research should focus on integrating pathological insights with innovative delivery technologies and personalized medicine approaches. With continuous advancements in delivery system design and deepening understanding of PDAC biology, nucleic acid therapeutics hold promise for overcoming the limitations of conventional therapies.

Authors contribution

Cong H: Writing-original draft, software, data curation.

Chang L: Writing-review & editing, software, data curation.

Li Y, Zou WB: Writing-review & editing, validation.

Xu C: Writing-review & editing, validation, funding acquisition.

Conflicts of interest

Congcong Xu is a Junior Executive Editor of BME Horizon. The other authors declare no conflicts of interest.

Ethical approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data and materials

Not applicable.

Funding

This work was supported by the National Natural Science Foundation of China (Grant No. U25C2006), Jiangsu Funding Program for Excellent Postdoctoral Talent (Grant No. 2025ZB320), Academic Degree and Postgraduate Education Reform Project of Jiangsu Province (Grant No. KYCX24_3300), Science and Technology Program of Suzhou City (Grant No. SYW2025195), Jiangsu Province Youth Science and Technology Talent Support Program (Grant No. JSTJ-2025-128), Chronic Disease Management Research Project of National Health Commission Capacity Building and Continuing Education Center (Grant No. GWJJMB202510022107), GenScript Life Science Research Grant Program and Natural Science Foundation of Jiangsu Province (Grant No. SBK20250405084).

Copyright

© The Author(s) 2026.