mRNA and lipid nanoparticle manufacturing

mRNA manufacture begins with a plasmid DNA template (covered in the plasmid record), which is linearised and transcribed in vitro by T7 RNA polymerase with a cap analogue and modified nucleotides. The RNA is purified by tangential flow filtration and chromatography (oligo-dT affinity, and reverse-phase or cellulose steps that remove the double-stranded RNA that triggers unwanted inflammation), then mixed with an ionisable lipid, a phospholipid, cholesterol and a PEG-lipid in a microfluidic or T-junction mixer so lipid nanoparticles self-assemble around it. The particles are buffer-exchanged, sterile-filtered, filled and frozen. Every step is cell-free, which is why COVID-19 vaccines could be scaled in months, and the same enzymes, cap analogues, lipids and single-use mixers are the supply chain for cancer vaccines.

Personalised neoantigen vaccines change the logistics. For intismeran autogene (mRNA-4157, Moderna with Merck) the tumour and blood are sequenced, up to thirty-four neoantigens are chosen by algorithm, a patient-specific plasmid is made and the RNA and nanoparticle steps are run as a single-patient GMP batch, with a turnaround of weeks; Moderna produces these at its Norwood, Massachusetts, site and has built a dedicated individualised therapy facility in Marlborough. BioNTech makes autogene cevumeran in Mainz on a similar cycle. Constraints are ionisable lipid patents and supply, enzyme and cap analogue suppliers, and the per-patient release testing that has to be as fast as the manufacture. A Phase 3 trial in melanoma (with pembrolizumab) and trials in lung and pancreatic cancer will decide whether this factory model is worth industrialising.