Why Circular RNA Is Redefining the RNA Therapeutic Landscape
Circular RNA outperforms linear mRNA in one fundamental respect: its covalently closed structure, which lacks 5′ caps and poly(A) tails, renders it resistant to exonuclease degradation. This topology is consistently cited across patent filings and academic literature as the central pharmacological advantage of circRNA — enabling prolonged protein expression and enhanced stability in vivo that linear mRNA cannot match without chemical modification.
The circRNA field has been accelerated by two converging forces: the success of COVID-19 mRNA vaccines demonstrating rapid clinical translation of RNA therapeutics, and growing awareness of linear mRNA’s limitations — particularly instability and immunogenicity. CircRNAs arise from backsplicing events mediated by the spliceosome, fusing a downstream splice donor to an upstream splice acceptor to produce covalently closed loops. This molecular architecture has attracted sustained attention from academic institutions, commercial biotechs, and national research institutes across three continents.
Backsplicing is the spliceosome-mediated process by which a downstream splice donor is joined to an upstream splice acceptor, producing a covalently closed circular RNA loop. This topology distinguishes circRNAs from canonical linear transcripts and is the structural basis for their resistance to exonuclease degradation.
According to WIPO, RNA therapeutics represent one of the fastest-growing patent categories in the life sciences, and circRNA is emerging as a distinct sub-class with its own IP architecture. The innovation signals mapped here span protein expression platforms, vaccine constructs, miRNA sponges, and circular prodrug nucleic acids — each representing a different engineering approach to exploiting circRNA’s structural advantages.
Circular RNA (circRNA) has a covalently closed structure lacking 5′ caps and poly(A) tails, rendering it resistant to exonuclease degradation — a property consistently cited as the central pharmacological advantage of circRNA over linear mRNA, enabling prolonged protein expression in vivo.
Four Therapeutic Modalities Shaping the circRNA Pipeline
The circRNA therapeutic pipeline clusters into four distinct modalities, each exploiting the molecule’s closed topology in a different way: translatable protein expression, vaccine antigen delivery, miRNA sponging for gene silencing, and circular prodrug nucleic acids (CPNs) for RNAi/ASO-mediated silencing.
1. Therapeutic Protein Expression via IRES-Mediated Translation
This is the dominant modality in the dataset. CircRNAs are engineered to encode therapeutic proteins via cap-independent translation initiated by internal ribosome entry site (IRES) elements. Retrieved results from both patents and academic papers describe IRES elements derived from Anabaena permuted intron-exon (PIE) systems, echovirus 29 (E29), and engineered group I intron fragments (T4Td) as translation initiation mechanisms. Key design features include group I intron fragments for autocatalytic circularization, duplex-forming homology arms to improve ligation efficiency, and purification strategies to remove linear RNA contaminants.
Constructs have been demonstrated encoding chimeric antigen receptor (CAR) T-cell targeting sequences, cytokines for intratumoral immune modulation, turboGFP and RBD reporters, and collagen VII for gene replacement therapy. Both Suzhou CureMed’s cmRNA platform and CirCode Biomedicine’s scalable circRNA system validate, in head-to-head comparisons, that IRES-driven circRNA produces stronger and more durable protein expression than linear mRNA in eukaryotic cells.
2. CircRNA Vaccines for Infectious Disease and Cancer
Multiple circRNA vaccine programs against SARS-CoV-2 and cancer have been documented. CircRNA vaccines leverage extended antigen production duration compared to 1-methylpseudouridine (1mΨ)-modified linear mRNA, and demonstrate Th1-skewed immune responses including potent neutralising antibodies, CD4+ T-cell activation, and IFN-γ+ CD8+ T-cell responses in murine and macaque models. For cancer vaccine applications, circRNA-LNP systems have been evaluated in three tumor models with evidence of adaptive immune activation, including combination with adoptive cell transfer approaches.
“CircRNA vaccines leverage extended antigen production duration compared to 1mΨ-modified linear mRNA, demonstrating Th1-skewed immune responses including potent neutralising antibodies and IFN-γ+ CD8+ T-cell responses in murine and macaque models.”
Mini-circRNA constructs from Virginia Commonwealth University employ rolling circle translation (RCT) to generate concatemeric antigen peptides, activating endosomal and cytosolic immune sensors as built-in adjuvants — a platform distinct from IRES-based approaches, relying on minimal non-coding sequences. Virginia Commonwealth University has filed four patent applications across WO, CA, AU, and US jurisdictions (2022–2024) covering this approach.
3. miRNA Sponges and Gene Silencing Decoys
Artificial circRNA constructs engineered to sequester specific miRNAs effectively de-repress their downstream mRNA targets. Key demonstrations include circular RNA decoys against miR-122 (inhibiting hepatitis C virus propagation), miR-21 (oncogenic/pro-proliferative miRNA, slowing tumor growth in 3D culture and xenograft models upon systemic delivery), and broad circRNA–miRNA sponge networks in cancer biology. The Aarhus University EP patent explicitly claims circular nucleic acid sequences designed to block miRNA interactions with RNA targets and act as RNA blockers or protein decoys.
4. Circular Prodrug Nucleic Acids (CPNs) for RNAi/ASO Gene Silencing
Arnay Sciences LLC filed an AU patent describing CPNs — oligonucleotides maintained in circular form until cleaved intracellularly by RNase H, Dicer, or other endogenous factors to release functional antisense or siRNA payloads. This approach targets a broad array of disease-relevant transcripts including PCSK9 (hypercholesterolaemia), TTR (transthyretin amyloidosis), DMD (Duchenne muscular dystrophy), SOD1 (ALS), ANGPTL3, APOC3, and MAPT. Separately, Alnylam Pharmaceuticals describes small circular interfering RNAs (sciRNAs) synthesised via click chemistry cyclization and conjugated to trivalent GalNAc for liver-targeted delivery as potent hepatic gene-silencing agents.
Circular prodrug nucleic acids (CPNs), as described in the Arnay Sciences LLC AU patent, are oligonucleotides maintained in circular form until cleaved intracellularly by RNase H, Dicer, or other endogenous factors to release functional antisense or siRNA payloads targeting transcripts including PCSK9, TTR, DMD, SOD1, ANGPTL3, APOC3, and MAPT.
Explore the full circRNA patent landscape and identify white-space opportunities with PatSnap Eureka.
Search circRNA Patents in PatSnap Eureka →Patent Landscape: Who Controls the Foundational IP
Massachusetts Institute of Technology (MIT) is the most active single patent filer in this dataset, with at least six distinct patent records — active or pending — in Israel and Singapore jurisdictions, all centring on group I intron-based circRNA platforms encoding CARs, antigens, and therapeutic proteins. MIT’s filings span 2020–2025, indicating sustained and expanding IP prosecution across the foundational circRNA production method space.
Orna Therapeutics holds two Singapore patent filings (pending, 2022) describing circRNA platforms with multiple expression sequences separated by cleavage sites, incorporating post-splicing group I intron fragments. Orna’s filings are structurally related to MIT’s foundational work, suggesting a licensing or spinout relationship. Drug developers entering this space should conduct freedom-to-operate analysis with respect to permuted intron-exon, ribozyme-mediated, and splint-ligation circularization methods before advancing any IRES-driven circRNA platform, according to PatSnap’s freedom-to-operate analysis capabilities.
CureVac SE has filed two patent applications (WO 2023, US 2025, pending) describing improved circular RNA constructs incorporating translation initiation sequences, coding sequences, UTR sequences, and poly(A) sequences for therapeutic protein expression, with vaccine applications explicitly claimed. Beijing Changping Laboratory — a Chinese government-affiliated research entity — filed a US patent (pending, 2025) on naked circRNA administration encoding therapeutic polypeptides including antigens, functional proteins, and targeting proteins such as antibodies, indicating Chinese government-affiliated research entering IP prosecution phase.
MIT holds a cluster of active and pending patents on group I intron-based circRNA production across multiple jurisdictions (IL, SG), with Orna Therapeutics holding structurally related IP. Drug developers entering this space should conduct freedom-to-operate analysis with respect to permuted intron-exon, ribozyme-mediated, and splint-ligation circularization methods before advancing any IRES-driven circRNA platform.
TherORna Inc. filed a WO patent (2025) on circular RNA-based therapeutic vaccines targeting solid tumors, explicitly incorporating MHC presentation pathway design with immune enhancer elements flanking antigen coding sequences. The European Patent Office and USPTO are both active jurisdictions for circRNA prosecution, with WO filings indicating international protection strategies from multiple assignees.
Massachusetts Institute of Technology (MIT) is the most active single circRNA patent filer in the dataset, with at least six distinct patent records in Israel and Singapore jurisdictions spanning 2020–2025, all centring on group I intron-based circRNA platforms encoding chimeric antigen receptors (CARs), antigens, and therapeutic proteins.
Translational Signals: From Mouse Models to Macaques
The circRNA dataset is predominantly preclinical in nature, but several non-human primate (NHP) studies provide meaningful IND-enabling signals for the vaccine modality specifically. No retrieved result explicitly references a clinical trial enrollment, IND filing, or regulatory submission for a circRNA-specific therapeutic or vaccine at the time of these publications.
Multiple retrieved papers describe circRNA RBD vaccine studies in rhesus macaques, providing immunogenicity and protection data. Studies from NIFDC (Beijing) and Wenzhou Medical University both report robust neutralising antibody responses and T-cell immunity in macaques following intramuscular circRNA-LNP administration. The circRNA RBD-Omicron and RBD-Delta constructs demonstrate cross-variant protection signals — an important translational finding given the ongoing evolution of SARS-CoV-2. These results were corroborated by independent research groups working with trimeric RBD configurations, assessed in both murine and NHP models.
On the delivery side, Suzhou CureMed describes a biodegradable ionizable lipidoid (AX4-LNP) formulation for circular mRNA delivery against SARS-CoV-2 Delta variant, with reported faster hepatic and splenic clearance compared to SM-102-based commercial LNPs — a translational safety signal relevant to regulatory submissions. According to FDA guidance on RNA-based therapeutics, LNP safety profiles and biodistribution are key IND-enabling data requirements, making this finding strategically significant.
Tsinghua University reports anti-tumor efficacy of circRNA-LNP in three murine tumor models, including a combination with adoptive cell transfer — representing early-stage IND-enabling data for the cancer vaccine modality. In the gene silencing space, the miR-21 sponge circRNA developed at the University of Giessen reduced tumor growth in xenograft models upon systemic delivery, providing proof-of-concept for circRNA-based oncology gene silencing.
The broader RNAi therapeutic landscape provides a relevant clinical validation precedent: siRNA-based drugs including patisiran, givosiran, and lumasiran have achieved regulatory approval, as noted by NIH-affiliated researchers. CircRNA-based gene silencing approaches via CPNs and sciRNAs seek to improve upon these established modalities by importing circularization as a stability-enhancing modification onto established RNAi/ASO chemistry frameworks — potentially providing a faster regulatory path than entirely novel circRNA modalities.
Track circRNA clinical and preclinical developments in real time using PatSnap Eureka’s drug pipeline intelligence.
Explore Drug Pipeline Data in PatSnap Eureka →Emerging Directions: Combination Strategies and Platform Engineering
Beyond the four core modalities, retrieved results signal several emerging combinatorial and platform-level directions that are shaping competitive differentiation in the circRNA space.
Immunogenicity Engineering: The Self vs. Foreign Axis
Stanford University’s 2017 paper on RIG-I sensing of foreign circRNA is a foundational result informing current platform design: use of human intron sequences in the circularization element abrogates innate immune activation, while foreign-intron circRNAs potently activate RIG-I. This creates a deliberate design axis between “low-immunogenicity” circRNA for therapeutic protein expression and gene therapy (where innate immune activation is undesirable) and “self-adjuvanted” circRNA for vaccines (where immune activation is the goal). Platform developers and investors should evaluate assets against this axis — the underlying circularization method and intron identity determine which regulatory and immunological framework applies.
Clean-PIE and Scarless Circularization
Multiple 2022 papers from Suzhou CureMed describe the Clean-PIE strategy — permuted T4Td intron-mediated ligation — for generating scarless circRNAs with minimised residual immunogenicity. This addresses a known limitation of the Anabaena PIE platform. Manufacturing process purity and immunogenicity reduction are cited as competitive differentiators in the circRNA platform space, suggesting that manufacturing IP complements composition-of-matter claims and may be independently licensable.
Computational Design: The circDesign Algorithm
StemiRNA Therapeutics (Shanghai) published the circDesign computational circRNA design algorithm (2023), optimising folding at each segment to enhance circularization efficiency, stability, and translatability simultaneously. The algorithm was demonstrated with rabies virus glycoprotein (RABV-G) as a vaccine antigen model, signalling that computational rational design is becoming integral to circRNA platform engineering. This development, tracked via PatSnap’s R&D intelligence tools, reflects a broader trend of in silico optimisation accelerating RNA therapeutic development.
Exosome-Mediated Delivery and CAR-T Applications
Retrieved results cite exo-circRNAs — circRNAs packaged in exosomes — as an anticancer delivery paradigm, including tumor suppression and diagnostic applications. This represents an early-stage but distinct delivery vector being explored in parallel with LNP systems. MIT’s patents consistently list CAR-encoding as a primary embodiment, and Suzhou CureMed’s EP patent mentions CAR-T cell therapy as an application. Signals suggest circRNA’s prolonged expression profile is being evaluated as an alternative to viral transduction for transient CAR expression in T-cell engineering.
Stanford University’s research on RIG-I sensing of foreign circular RNA established that use of human intron sequences in the circularization element abrogates innate immune activation, while foreign-intron circRNAs potently activate RIG-I — a finding that informs the deliberate design split between low-immunogenicity therapeutic circRNAs and self-adjuvanted vaccine circRNAs.
Strategic Implications for Drug Developers and Investors
The circRNA innovation landscape presents both significant opportunities and specific risk factors that drug developers and investors need to assess before committing resources to this modality.
SARS-CoV-2 as proof-of-concept, not the terminal indication. The density of circRNA vaccine data against SARS-CoV-2 RBD reflects the COVID-19 era’s role as an accelerant for circRNA development. Signals from retrieved results suggest cancer vaccines, CAR-T manufacturing, and intratumoral cytokine delivery are the primary near-term commercial development targets, where circRNA’s durability advantage over linear mRNA has greater clinical impact.
Manufacturing differentiation is a key competitive moat. Retrieved results consistently identify in vitro circularization at industrial scale as a primary technical barrier. Platform developers who demonstrate scalable, scarless, low-immunogenicity circRNA synthesis — such as the Clean-PIE and co-transcriptional self-catalysis approaches — hold manufacturing IP that complements composition-of-matter claims and may be independently licensable. The ISO standards framework for RNA manufacturing quality is still evolving, creating an early-mover advantage for organisations that establish robust process documentation.
Gene silencing via circRNA remains at proof-of-concept stage. In contrast to protein expression and vaccine applications (which have NHP data), circRNA-based miRNA sponging and CPN-mediated gene silencing are represented primarily by cell culture and murine xenograft data in this dataset. The sciRNA (Alnylam) and CPN (Arnay Sciences) approaches import circularization as a stability-enhancing modification onto established RNAi/ASO chemistry frameworks — potentially providing a faster regulatory path than entirely novel circRNA modalities.
“Platform developers who demonstrate scalable, scarless, low-immunogenicity circRNA synthesis hold manufacturing IP that complements composition-of-matter claims and may be independently licensable.”
Dual-use immunogenicity design axis creates platform bifurcation. Retrieved results reveal a deliberate split between low-immunogenicity circRNA (for protein replacement and gene therapy) and self-adjuvanted circRNA (for vaccines). The underlying circularization method and intron identity determine which regulatory and immunological framework applies — a distinction that must be reflected in both IP strategy and clinical development planning.
Note: This analysis is derived from a limited set of patent and literature records retrieved across targeted searches. It represents a snapshot of innovation signals within this dataset only and should not be interpreted as a comprehensive view of the full field, clinical pipeline, or regulatory landscape.