Advancing circRNA Therapeutics: Tackling Development And Manufacturing Challenges

A conversation between Circio’s CEO, Erik Wiklund; and Life Science Connect’s Michelle Raley

Circular RNA (circRNA) is gaining attention as an emerging RNA modality with the potential to provide more durable gene expression than conventional linear mRNA. Its circular structure can enhance RNA stability, creating opportunities for prolonged protein expression and potentially expanding its utility across gene and cell therapy applications. At the same time, questions remain around delivery, manufacturing, and analytical characterization.

In this Q&A, Life Science Connect’s Michelle Raley catches up with Erik Wiklund, CEO of Circio, to discuss the advantages of circRNA, the hurdles to tackle, Circio’s unique vector-based approach and the manufacturing and delivery challenges facing the field, and where the opportunities exist for advancing this modality.

Circular RNA is often discussed as a next-generation RNA modality. From a scientific perspective, what do you believe are the most important advantages circRNA offers over linear RNA approaches?

circRNA has no free ends, so it is not degraded by the exonucleases that digest linear mRNA transcripts. With our circVec platform, we observe around 75-fold longer half-life for circRNA produced in vivo compared to mRNA. This leads to prolonged and increased protein expression, which is a major advantage in a therapeutic setting. This is important because it can enable more effective, safer, and lower-cost next-generation gene and cell therapies to patients.

In the therapeutic context, higher expression per vector allows a lower dose. This is particularly relevant for adeno-associated virus (AAV) gene therapy, where dose is linked to both safety and cost of goods. In vivo, we consistently observe an expression advantage of around 50-fold for AAV vectors using our circVec circular RNA system compared to conventional mRNA-based expression. This advantage has so far been shown in heart, eye, and CNS. The effect is consistent across multiple tissues, cell types, and AAV serotypes, which gives us confidence that the advantage of circular RNA-based gene expression is a broadly applicable biological feature.

Another therapeutic area where safety and cost of goods are an issue is cell therapy. While autologous CAR T-cell therapies have shown deep and durable responses in certain liquid cancers, they are highly complex to manufacture and difficult to scale, in effect generating a unique, individualized drug for each patient. In recent years, there has therefore been a push to move to in vivo CAR T-cell strategies, where a viral vector or lipid nanoparticle (LNP)-formulated RNA is used to express the CAR constructs directly in the patient’s T cells. While this has the advantage of being a more scalable approach, the most advanced approaches are lentiviral and therefore have serious safety concerns, as they are genome-integrating and can lead to secondary malignancies.

circRNA is an attractive alternative to lentiviral vectors, as it does not carry the risk of genome integration. Furthermore, non-viral delivery approaches can enable redosing, giving more flexibility. circRNA also shows an advantage over linear mRNA approaches in this setting, and with circVec, we have seen over six months’ expression in immune tissues compared with only days for LNP-delivered mRNA and circRNA.

circRNA is not superior in every setting, as circRNA stability varies between cell types. As such, the advantage can vary depending on the target tissue, and indications should be chosen with this in mind. We observe that circVec is a particularly effective gene expression system in heart, eye, and CNS. Conversely, circRNA is less stable in the liver, which makes this a less suitable organ for circRNA-based therapeutics. However, this liver effect can also be an advantage in many settings, as it reduces off-target expression that can be a major cause of liver toxicity.

As the industry moves from proof-of-concept studies toward clinical development, what do you see as the most significant manufacturing challenges that must be solved to enable scalable and reproducible circRNA production? And what first steps of progress are you seeing from industry to tackle these challenges?

The challenges differ for synthetic circRNA and for DNA-encoded circRNA. For synthetic circRNA, the main bottlenecks are circularization efficiency and purification. Synthesis relies on ligases to close the RNA into circles, but this is often accompanied by by-products, which reduces synthesis efficiency and makes purification and recycling more difficult.

With circVec, DNA, not RNA, is the starting point. Therefore, we do not face the same manufacturing challenges that have slowed the implementation and clinical entry of in vitro transcribed (IVT) circRNA. In Circio’s case, it is the vector chosen that defines the specific manufacturing implications. For example, by inserting our circVec expression system into AAV vectors, the genetic payload will be expressed at higher levels than for comparable mRNA-based vectors. This increased potency in turn allows for dose reduction, as you can achieve the same therapeutic index with fewer viral particles delivered. This improves the safety for the patient and also decreases the amount of vector that has to be produced. Therefore, switching from standard mRNA-based AAV to a circular RNA expression strategy can both simplify and lower the cost of gene therapy manufacturing.

circRNA-based gene expression can also be applied to non-viral delivery approaches. Circio is working on LNP-formulated synthetic DNA for in vivo CAR T-cell applications. In this program, we are collaborating with several partners on linear and circular single-stranded DNA vectors designed to be less immunostimulatory. Single-stranded DNA is still a new format, there is no standardized manufacturing process yet, and it has never been produced at scale. By working with several partners in parallel, we reduce our dependence on any single technology and de-risk this part of the program.

The broader RNA field continues to refine analytical methods and quality benchmarks. What advances are still needed in circRNA characterization and quality control to support regulatory confidence and commercial readiness?

Circio does not deliver circular RNA itself. Having discovered human circRNA and described how these are generated in vivo, we used our first-mover and discoverer position and moved beyond pure circular RNA to deliver DNA that encodes circRNA. Through this in vivo approach, we can trick the patient’s own cells to express a circRNA of our design. The advantage of this approach is that Circio’s circRNAs are endogenously produced and viewed as fully natural by the cell. In addition, the vector-based delivery ensures a constant supply of newly generated circRNA, which substantially prolongs the therapeutic durability. Detailed characterization of the circRNA molecule, such as its purity and structural integrity, is therefore mainly a question for companies that deliver synthetic circRNA, and not relevant for circVec.

For our approach, quality control starts from the established frameworks for AAV and DNA vectors, covering identity, purity, vector genome content, and potency. We are an early-stage company and are still defining the full analytical package, together with manufacturing partners and, in time, regulators.

How does the unique biology of circular RNA influence discussions around delivery, durability of expression, and therapeutic performance, and what key questions remain unanswered in these areas?

For synthetic circRNA, delivery constraints are largely the same as for linear mRNA, since it is a large, highly charged RNA that depends on a delivery vehicle. For circVec delivered by AAV, the defining feature is the very strong expression we observe. This has two consequences. First, the therapeutic window needs to be characterized, because the expression level that provides the required benefit and associated toxicity will vary between transgenes and tissues. Second, circVec allows a significantly lower vector dose to reach a given level of expression compared to conventional AAV, which matters for both safety and cost of goods, as AAV safety concerns are largely dose-dependent.

For non-viral delivery, our focus is DNA delivery of circVec using immune-silent DNA. LNPs and other vehicles have mainly been optimized for RNA, whereas DNA is generally larger, has to reach the nucleus to be transcribed, and can trigger innate DNA sensors if it is exposed in the cytosol. Finding vehicles that deliver immune-silent DNA efficiently, with minimal immune activation, is therefore a priority for this part of the platform.

Several questions remain open, and the most immediate is how much expression is required and safe in a given therapeutic situation. At some level, protein folding and secretory capacity probably become limiting. An overloaded secretory pathway activates the unfolded protein response (UPR), which can impair cell function and, if prolonged, lead to cell death. With circVec, the build-up of both RNA transcripts and protein expression is slower than when using the mRNA route. As a consequence, we consistently observe reduced activation of the UPR pathway, which suggests that circRNA-based expression of a genetic cargo may in fact be less stressful for the cell.

The mRNA field has experienced tremendous growth over the past several years. What lessons from the development and manufacturing of mRNA therapeutics do you believe are most applicable to the advancement of circular RNA platforms over the next decade?

mRNA products shared a set of design, manufacturing, and delivery standards that converged with the arrival of COVID-19, combined with a substantial boost in funding to rapidly scale manufacturing and supply chains. circRNA is still a new modality, and a manufacturing gold standard has yet to emerge. As the standard is set and broadly adopted through the industry, we predict that circRNA manufacturing will be lower cost than mRNA, as cost-driving enzymatic steps and the requirement for modified nucleotides are eliminated. In addition, the improved stability of the circular format suggests that the shelf life and storage logistics can also be simplified compared to more chemically unstable mRNA.

Applications should be chosen where the property of the platform matters. The design and manufacturing flexibility of mRNA is a major advantage for vaccines. The strength of IVT circRNA is enhanced durability, so the most promising applications are those where prolonged expression vs. mRNA is important. For circVec, we take the durability advantage one step further, applying it to generate more potent AAV gene therapies that can be delivered at low therapeutic doses, and a durable, non-integrating approach for in vivo cell engineering.

The circRNA field is rapidly accelerating and has attracted substantial investments, even before any clinical proof of concept has been generated. Lilly's acquisition of Orna, Bristol Myers Squibb's acquisition of Orbital, and Johnson & Johnson's partnership with Sail, along with the associated multibillion-dollar valuations, demonstrate the potential the industry sees in circRNA as a next-generation therapeutic format with a broad range of potential applications. Several programs are now in, or entering, the clinic, and there is an exciting period ahead for this emerging field as data start to emerge.

About The Expert:

Erik Wiklund, CEO of Circio, has deep scientific knowledge in RNA biology and was the co-discoverer of human circular RNA. He has 15 years of pharma and biotech industry experience in a variety of functions including R&D, finance, and business development. He previously worked for the radiopharmaceutical company Algeta, which was acquired by Bayer in 2014, and as a consultant in the Pharma & Health Care practice of McKinsey & Company. Wiklund holds a Ph.D. in molecular biology from Aarhus University, Denmark, and the Garvan Institute of Medical Research, Sydney, Australia.