What does it take to develop and manufacture a mRNA / saRNA drug?
An mRNA or self-amplifying RNA program is a chain of fairly distinct capabilities, and the modality rewards teams that know where one specialist's work ends and the next begins. It starts at the bench with sequence and construct design: codon optimization, the untranslated regions that drive translation, the cap and poly-A strategy, and for saRNA the replicase machinery that lets a smaller dose self-amplify inside the cell. Most of this early design and screening work is research-grade, run by a discovery CRO or in-house, and it sets the quality of everything downstream.
From there the work moves into CMC, which is where mRNA differs most from a small molecule or an antibody. The drug substance is made by in vitro transcription (IVT): an RNA polymerase reads a linearized DNA template, the RNA is capped (co-transcriptionally with a cap analog or enzymatically) and tailed, then purified to strip out template DNA, double-stranded RNA byproducts, residual NTPs, and enzyme. That plasmid template is a manufactured starting material in its own right, frequently the real item on the critical path. Naked RNA degrades in minutes and barely crosses a cell membrane, so almost every program pairs the RNA drug substance with LNP encapsulation (an ionizable lipid, a helper phospholipid, cholesterol, and a PEG-lipid mixed with the payload) as the drug-product step.
This is where specialist mRNA / saRNA CDMOs separate from generalists. A shop that runs CHO antibody campaigns every week is not set up for IVT, oligo-dT affinity capture, or dsRNA control, and the analytics are their own discipline: RNA integrity by capillary gel electrophoresis, capping efficiency by LC-MS, poly-A tail characterization, dsRNA quantitation, residual template DNA, and sequence confirmation. saRNA adds its own wrinkles, since the construct is longer and the replicon has to stay intact. RNA is one of the faster modalities to manufacture once the upstream pieces are ready, which is exactly why it carried the COVID vaccines, but that speed assumes the template, the cap strategy, and the validated methods already exist. Treating LNP and analytics as an afterthought is the usual reason an otherwise fast program stalls between suppliers.
How do you choose a CRO or CDMO for mRNA / saRNA?
The honest filter is fit to your construct and your stage, not the size of the facility. A vaccine-scale mRNA CDMO can be over-built for a rare-disease program that needs a few grams, and a supplier strong on standard mRNA may be a beginner at self-amplifying or circular RNA. Score two or three suppliers against the same written scope (sequence type, quantity, quality grade, cap strategy, whether LNP is included), and weigh the analytical and regulatory track record at least as heavily as the per-gram price. The checklist below covers the attributes that actually separate a clean engagement from a painful one.
- Relevant platform and track record: real, completed work in your specific construct (standard modified-nucleotide mRNA, saRNA, circular RNA) and your application (vaccine, protein replacement, in vivo editing), not a generic RNA claim. saRNA in particular needs a supplier who has handled the longer transcript and the replicon.
- GxP grade and analytical capability for this modality: confirm the actual grade on offer (research, non-GMP engineering, or full GMP under 21 CFR Part 211 or EU GMP) and whether the RNA-specific methods (integrity, capping efficiency, dsRNA, residual DNA template, poly-A tail) are qualified or validated rather than improvised per batch.
- Capacity and scale: a scale you can grow into, from milligram tox batches through gram-scale clinical supply, with a realistic slot. Ask where the critical path actually sits, usually template linearization, cap-analog and raw-material lead times, and method readiness rather than the IVT reaction itself.
- LNP and drug-product scope: whether the supplier encapsulates to LNP and fill-finishes in-house or stops at purified drug substance, and how the plasmid template is sourced and qualified, so you know exactly where the handoffs occur and whether a tech transfer is coming.
- Regulatory experience: which agencies the site has filed with (FDA, EMA, PMDA), recent inspection history, raw-material traceability, and whether they can author the CMC sections your IND needs. The test article in your GLP tox study has to match the clinical formulation, so the process should be locked before the pivotal studies start.
- IP and confidentiality: who owns process improvements made on your program, freedom-to-operate on any licensed capping or proprietary-lipid technology you would be locked into, and clean tech-transfer terms if you move the process later.