What RiboX's FDA clearance means for anyone who buys — or makes — IVT reagents

On August 8, 2026, RiboX Therapeutics announced FDA IND clearance for RXIM002 — the world's first circular RNA-based in vivo CAR-T therapy. Delivered via a targeted lipid nanoparticle (tLNP), RXIM002 encodes an anti-CD19 chimeric antigen receptor on a circRNA backbone, generating CAR-T cells inside the body without ex vivo manufacturing or lymphodepletion. The Phase 1 trial, POPULUS-1, will initially enroll patients with relapsed or refractory immune thrombocytopenia (ITP).

The headlines will, understandably, focus on the clinical first. They should. But for anyone who procures, manufactures, or develops the upstream reagents that make RNA therapeutics possible, RXIM002 quietly does something else: it rewrites the raw-material shopping list.

That matters because the post-COVID mRNA reset has taught the industry a hard lesson. Betting on a single topology is risky. Moderna, BioNTech, and CureVac have all restructured their pipelines as the respiratory vaccine market normalized. Meanwhile, RNA itself has fragmented into a family of architectures — linear mRNA, self-amplifying RNA (saRNA), and now circular RNA. Each has different strengths, but they share a surprising amount of upstream DNA. The durable investment, and the durable supply relationship, is the layer beneath the topology.

circRNA Is Not Just Another mRNA

Circular RNA is exactly what it sounds like — a covalently closed loop with no 5' cap and no 3' poly(A) tail. That topology changes the rules.

Without free ends, circRNA resists exonuclease degradation. Translation is cap-independent, driven by internal ribosome entry sites (IRES) or synthetic aptamers. The durability advantage is real and quantified: in the foundational work by Wesselhoeft et al. (2018, Nature Communications), circRNA produced 811% more protein than unmodified linear mRNA at 24 hours in HEK293 cells, and its protein-production half-life roughly doubled — approximately 80 hours in HEK293 and 116 hours in HeLa cells, versus 43-49 hours for linear mRNA.

Subsequent work by Chen et al. (2023, Nature Biotechnology) showed that optimizing five vector elements improved circRNA protein yields by several hundred-fold, with more durable in vivo translation. For a modality whose clinical promise depends on durable protein expression — whether a CAR receptor on T cells or a replacement enzyme in a target tissue — that extra translational half-life is not a marginal gain. It is the reason a single, transient RNA dose can behave like a sustained therapy.

The molecule works. The question is what it takes to make it.

The Bill of Materials, Rewritten

Here is the insight that almost no one is discussing: a cap-free, tail-free circRNA still runs on the same IVT engine as linear mRNA — it just sheds the capping and polyadenylation stack.

Figure 1. The RNA bill of materials, rewritten: linear mRNA IVT vs. circRNA PIE self-splicing. Green = retained in both topologies; orange = circRNA-specific; gray = eliminated in circRNA.

Both modalities start identically: a DNA template, linearized by restriction enzymes such as BsaI or BspQI, transcribed by T7 RNA polymerase with NTP substrates. Both require DNase I for template removal, RNase inhibitor to protect the transcript, and inorganic pyrophosphatase to drive yield forward. These are not niche inputs; they are the backbone of every commercial IVT reaction, whether the output is a vaccine, an oncology therapeutic, or a CAR-encoding circRNA.

Where they diverge:

Still essential for circRNA: T7 RNA polymerase; restriction enzymes for linearization; natural NTPs, with extra GTP as the PIE splicing cofactor; DNase I; RNase inhibitor; pyrophosphatase; RNase R for circularization enrichment; and — critically — dsRNA and purity QC.

Eliminated for circRNA: Capping enzymes (vaccinia capping enzyme, 2'-O-methyltransferase), cap analogs, poly(A) polymerase, capping-efficiency assays, and poly(A) tail-length analysis. None of these apply to a molecule with no cap and no tail.

For procurement and manufacturing teams, the math is striking. A circRNA program strips out the most expensive and IP-encumbered corner of the linear mRNA reagent list — the cap — while preserving the high-volume enzyme layer. The topology changes. The upstream toolkit does not.

A Counter-Intuitive Note on Modified Nucleotides

In linear mRNA, N1-methylpseudouridine (m1Ψ) is transformative — it dampens innate immunity and boosts translation. In circRNA, it can do the opposite. Wesselhoeft et al. (2019, Molecular Cell) demonstrated that, unlike linear mRNA, circRNA does not benefit from m1Ψ modification; it can actually impair IRES-driven translation. The lesson is humbling: modified nucleotides are not universal optimizers. What works for one topology may undermine another.

Figure 2. circRNA outperforms linear mRNA in protein output and durability. (A) Relative protein output at 24 h in HEK293 cells. (B) Protein-production half-life.

Purity Is Not a Bonus — It Is the Product

If there is one technical variable that determines whether circRNA succeeds or fails, it is purity.

The literature sends mixed signals. Wesselhoeft et al. (2019) reported that highly purified, unmodified circRNA can bypass RIG-I and TLR sensing, making it less immunogenic than linear mRNA. Chen et al. (2017, 2019, Molecular Cell) found that engineered circRNA activates RIG-I. Liu et al. (2022) showed that permuted intron-exon (PIE) derived extraneous fragments can trigger PKR.

These findings are not contradictory — they are purity-dependent. The immunogenicity of circRNA is not an intrinsic property of the molecule; it is a function of what is left in the vial after purification. Residual linear RNA, nicked circRNA, intron fragments from PIE splicing, triphosphorylated ends, and double-stranded RNA (dsRNA) impurities are the actual culprits. Remove them, and circRNA behaves. Leave them, and it lights up innate immunity.

The control strategy is therefore multi-layered: start with a low-dsRNA T7 RNA polymerase to reduce the dominant by-product; use RNase R digestion to enrich for circular species; polish with RP-HPLC or size-exclusion chromatography; and close the loop with a sensitive dsRNA assay. Each step is negotiable in early research; in a GMP workflow, none of them are optional.

This makes dsRNA quantification and impurity profiling not a nice-to-have but an entry ticket. The USP Analytical Procedures for mRNA Vaccine Quality guidelines have reached their third edition (August 2024), now retitled to include therapeutics — but there is no dedicated pharmacopeial chapter for circRNA. The regulatory landscape is still white space. Whoever controls impurities controls the modality.

Figure 4. circRNA purity: impurity sources, innate immune sensors, and control points. Immunogenicity is a purity-dependent property, not an intrinsic feature of the topology.

The In Vivo CAR-T Gold Rush

RXIM002 does not exist in a vacuum. It is part of a capital wave that has reshaped cell therapy.

The momentum is clear. From late 2024 through 2026, major pharmaceutical companies—including Lilly, AbbVie, Bristol Myers Squibb, AstraZeneca, and Gilead/Kite—announced multibillion-dollar deals targeting in vivo CAR-T platforms. Lilly acquired Kelonia for up to $7 billion, on top of its earlier Orna acquisition at up to $2.4 billion. AbbVie closed Capstan for up to $2.1 billion. Bristol Myers Squibb took Orbital for roughly $1.5 billion. AstraZeneca partnered with EsoBiotec for up to $1 billion. Gilead/Kite acquired Interius for $350 million and struck a deal with Pregene worth up to $1.64 billion.

Why autoimmune disease as the beachhead, not oncology? The answer is biology plus economics. Ex vivo CAR-T in cancer justifies its manufacturing complexity because the alternative is often fatal. In autoimmunity, the patient population is larger, the disease chronic, and the safety bar higher — transient CAR expression becomes a feature, not a limitation. Georg Schett's group at Erlangen demonstrated that CD19 CAR-T can induce drug-free remission in severe refractory lupus — first in a single patient (NEJM, 2021), then in a five-patient series (Nature Medicine, 2022), and most recently in a 15-patient case series across SLE, myositis, and systemic sclerosis (NEJM, 2024). Patients achieved DORIS remission, autoantibodies vanished, and vaccine responses were preserved even after naïve B cells returned. Schett's framing: "Not only are the patients in drug-free remission, but I believe many are cured."

For ITP specifically — RXIM002's first indication — the clinical need is acute. Adult prevalence sits at 9.5-23.6 per 100,000, and Mayo Clinic data suggest that as many as 80% of patients will eventually relapse and become steroid-refractory. Standard-of-care options (corticosteroids, IVIG, rituximab, TPO-RAs, splenectomy) leave a substantial gap.

The manufacturing implication is profound. In vivo CAR-T does not eliminate manufacturing burden — it relocates it. Every dose is an industrial-scale IVT run. The cost-of-goods moves from GMP clean rooms to RNA reactors, and the demand for high-quality IVT enzymes, substrates, and analytics scales accordingly.

Figure 3. In vivo CAR-T big-pharma acquisition wave (2024-2025). Values include upfront and milestone payments where disclosed.

The Market Signal

The GMP-grade IVT enzyme market — valued at approximately $362 million in 2024 and projected to reach $923 million by 2034 (CAGR 10.4%, InsightAce) — is one indicator of where the puck is going. But the more important signal is structural: RNA is diversifying into a portfolio of topologies, and every topology still depends on the same upstream manufacturing layer.

Figure 5. GMP-grade IVT enzyme market forecast (2024-2034). Source: InsightAce.

The Picks-and-Shovels Layer

This is where we come in.

At Hzymes, we have spent years building the upstream toolkit that RNA therapeutics run on — not for one modality, but for all of them. The circRNA revolution does not change our value proposition; it validates it. When the topology shifts from linear to circular, from capped to cap-free, from polyadenylated to IRES-driven, the enzyme layer underneath remains the same.

For circRNA developers, the products that matter are the ones that survive the BOM rewrite:

GMP T7 RNA Polymerase — including our low-dsRNA variant (Cat. HBP000340, 0.20% dsRNA, 3-5× lower than wild-type) and thermostable M16 variant (HBP000350, suitable for PIE Self-Splicing), all backed by FDA DMF MF036856. In a modality where dsRNA impurity is the swing factor, low-dsRNA enzymes are not a luxury — they are the first line of defense.

GMP Restriction Enzymes — BsaI (HBP001306-310) and BspQI (HBP001406), both with FDA DMF filings, for precise template linearization.

GMP Natural NTPs — ATP, GTP, CTP, UTP. Unmodified nucleotides are the circRNA standard; GTP additionally serves as the PIE splicing cofactor.

GMP RNase Inhibitor (HBP000402-406, DMF MF036854), Yeast Inorganic Pyrophosphatase (HBP000501-505, DMF MF036853), and Salt-Tolerant DNase I (HBP000910) — the IVT workhorses that every topology depends on.

High-Yield IVT Kits — achieving 9-11 g/L yields with templates up to 12 kb, because large circRNA constructs (CARs; dystrophin-scale precedents) demand long-template capability.

dsRNA ELISA QC Kit (HBP003800/3801, sensitivity to 0.1 pg/well) — because if you cannot measure dsRNA, you cannot control it.

One-Stop mRNA CRO Services — from sequence design to custom RNA/LNP synthesis (500 µg to gram scale), microfluidic LNP encapsulation, and a full analytics panel including dsRNA ELISA, residual DNA qPCR, endotoxin LAL, and integrity by capillary electrophoresis.

What ties these products together is not a single SKU. It is a manufacturing philosophy: GMP-grade, animal-origin-free (AOF), supported by FDA DMF filings, and produced at scale — with a 25 kg per-batch enzyme capacity and dual production centers in Shanghai and Wuhan. That matters for circRNA developers who are moving from bench to clinic and need a supplier that can grow with them.

We do not pretend that our capping enzymes, cap analogs, or poly(A) tail-length kits serve circRNA — they do not. What serves circRNA is the same T7/restriction-enzyme/NTP/pyrophosphatase/RNase-inhibitor backbone that has powered linear mRNA all along, plus the dsRNA QC and analytics that purity-dependent modalities demand.

The Bigger Picture

RNA is no longer a single molecule. It is a portfolio of topologies — linear, self-amplifying, circular — and the industry is diversifying precisely because no single topology de-risks everything. For drug developers, that means spreading bets across modalities. For suppliers, it means one thing: the layer that survives every topology shift is the one worth investing in.

RXIM002 proved that a circRNA can reach the clinic. The question now is whether the upstream supply chain can keep up — at GMP grade, at scale, with the purity standards that circRNA demands, and with the regulatory credentials that global buyers require.

We built Hzymes to be that layer.

References

1.     RiboX Therapeutics announces FDA IND clearance for RXIM002, the first circular RNA-based in vivo CAR therapy for autoimmune cytopenias [press release]. Shanghai and Cambridge, MA: PR Newswire; August 8, 2026. https://www.prnewswire.com/news-releases/ribox-therapeutics-announces-fda-ind-clearance-for-rxim002-the-first-circular-rna-based-in-vivo-car-therapy-for-autoimmune-cytopenias-302846525.html

2.     Wesselhoeft RA, Kowalski PS, Anderson DG. Engineering circular RNA for potent and stable translation in eukaryotic cells. Nat Commun. 2018;9(1):2629. doi:10.1038/s41467-018-05096-6

3.     Chen R, Wang SK, Belk JA, et al. Engineering circular RNA for enhanced protein production. Nat Biotechnol. 2023;41(2):262-272. doi:10.1038/s41587-022-01393-0

4.     Wesselhoeft RA, Kowalski PS, Parker-Hale FC, Huang Y, Bisaria N, Anderson DG. RNA circularization diminishes immunogenicity and can extend translation duration in vivo. Mol Cell. 2019;74(3):508-520.e4. doi: 10.1016/j.molcel.2019.02.015

5.     Chen YG, Kim MV, Chen X, et al. Sensing self and foreign circular RNAs by intron identity. Mol Cell. 2017;67(2):228-238.e5. doi: 10.1016/j.molcel.2017.05.022

6.     Chen YG, Chen R, Ahmad S, et al. N6-methyladenosine modification controls circular RNA immunity. Mol Cell. 2019;76(1):96-109.e9. doi: 10.1016/j.molcel.2019.07.016

7.     Liu CX, Guo SK, Nan F, Xu YF, Yang L, Chen LL. RNA circles with minimized immunogenicity as potent PKR inhibitors. Mol Cell. 2022;82(2):420-434.e6. doi: 10.1016/j.molcel.2021.11.019

8.     United States Pharmacopeia. Analytical Procedures for Quality of mRNA Vaccines and Therapeutics (draft guidelines, 3rd edition). Rockville, MD: USP; August 2, 2024. https://www.uspnf.com/notices/analytical-procedures-mrna-vaccines-20240802

9.     Vyriad announces strategic collaboration with Novartis to develop in vivo CAR-T cell therapies [press release]. Rochester, MN: PR Newswire; November 20, 2024. https://www.prnewswire.com/news-releases/vyriad-announces-strategic-collaboration-with-novartis-to-develop-in-vivo-car-t-cell-therapies-302310717.html

10.   Lilly to acquire Kelonia Therapeutics to advance in vivo CAR-T cell therapies [press release]. Indianapolis, IN: Eli Lilly and Company; April 20, 2026. https://investor.lilly.com/news-releases/news-release-details/lilly-acquire-kelonia-therapeutics-advance-vivo-car-t-cell

11.   Lilly to acquire Orna Therapeutics to advance cell therapies [press release]. Indianapolis, IN: Eli Lilly and Company; February 9, 2026. https://investor.lilly.com/news-releases/news-release-details/lilly-acquire-orna-therapeutics-advance-cell-therapies

12.   AbbVie to acquire Capstan Therapeutics, further strengthening commitment to transforming patient care in immunology [press release]. North Chicago, IL: AbbVie; June 30, 2025. https://news.abbvie.com/2025-06-30-AbbVie-to-Acquire-Capstan-Therapeutics,-Further-Strengthening-Commitment-to-Transforming-Patient-Care-in-Immunology

13.   Bristol Myers Squibb strengthens and diversifies cell therapy portfolio with acquisition of Orbital Therapeutics [press release]. Princeton, NJ: Bristol Myers Squibb; October 10, 2025. https://news.bms.com/news/details/2025/Bristol-Myers-Squibb-Strengthens-and-Diversifies-Cell-Therapy-Portfolio-with-Acquisition-of-Orbital-Therapeutics/default.aspx

14.   AstraZeneca to acquire EsoBiotec to advance cell therapy ambition [press release]. Cambridge, UK: AstraZeneca; March 17, 2025. https://www.astrazeneca.com/media-centre/press-releases/2025/astrazeneca-to-acquire-esobiotec.html

15.   Kite to acquire Interius BioTherapeutics to advance in vivo platform [press release]. Foster City, CA: Gilead Sciences; August 21, 2025. https://www.gilead.com/news/news-details/2025/kite-to-acquire-interius-biotherapeutics-to-advance-in-vivo-platform

16.   Kite puts $1.6B on the line to pair up with China's Pregene for another in vivo CAR-T deal. Fierce Biotech. October 17, 2025. https://www.fiercebiotech.com/biotech/kite-puts-16b-line-pair-chinas-pregene-another-vivo-car-t-deal

17.   Johnson & Johnson announces collaboration with Sail Biomedicines to advance in vivo CAR-T programs and transform autoimmune disease through immune reset [press release]. New Brunswick, NJ: Johnson & Johnson; July 29, 2026. https://www.jnj.com/media-center/press-releases/johnson-johnson-announces-collaboration-with-sail-biomedicines-to-advance-in-vivo-car-t-programs-and-transform-autoimmune-disease-through-immune-reset

18.   Mougiakakos D, Krönke G, Völkl S, et al. CD19-targeted CAR T cells in refractory systemic lupus erythematosus. N Engl J Med. 2021;385(6):567-569. doi:10.1056/NEJMc2107725

19.   Mackensen A, Müller F, Mougiakakos D, et al. Anti-CD19 CAR T cell therapy for refractory systemic lupus erythematosus. Nat Med. 2022;28(10):2124-2132. doi:10.1038/s41591-022-02017-5

20.   Müller F, Taubmann J, Bucci L, et al. CD19 CAR T-cell therapy in autoimmune disease — a case series with follow-up. N Engl J Med. 2024;390(8):687-700. doi:10.1056/NEJMoa2308917

21.   TIME100 Health: Georg Schett. TIME. May 2, 2024. https://time.com/collections/time100-health/6968938/georg-schett/

22.   Segal JB, Powe NR. Prevalence of immune thrombocytopenia: analyses of administrative data. J Thromb Haemost. 2006;4(11):2377-2383. doi:10.1111/j.1538-7836.2006. 02147. x

23.   Feudjo-Tepie MA, Robinson NJ, Bennett D. Prevalence of diagnosed chronic immune thrombocytopenic purpura in the US: analysis of a large US claim database: a rebuttal. J Thromb Haemost. 2008;6(4):711-712. doi:10.1111/j.1538-7836.2008. 02911. x

24.   Ghanima W, Gernsheimer T, Kuter DJ. How I treat primary ITP in adult patients who are unresponsive to or dependent on corticosteroid treatment. Blood. 2021;137(20):2736-2744. doi:10.1182/blood.2021010968

25.   InsightAce Analytic. GMP-Grade IVT Enzymes for Therapeutic RNA Market. Report ID 3248. Pune, India: InsightAce Analytic; 2025. https://www.insightaceanalytic.com/report/gmp-grade-ivt-enzymes-for-therapeutic-rna-market/3248