RNA vaccines have proven effective against COVID‑19 and are now being extended to diseases such as cancer, yet they require storage at –20 °C to –80 °C, limiting distribution. Researchers at MIT’s Koch Institute used an AI algorithm to fine‑tune the lipid nanoparticle (LNP) formulation, achieving stability at room temperature for up to a year and at about 100 °F (≈37 °C) for two months.
They first evaluated roughly 50 FDA‑approved excipients by incorporating each into LNPs and measuring protection of firefly‑luciferase mRNA via bioluminescence. Five top‑performing excipients were fed into the AI model, which predicted optimal ratios. Iterative cell‑based testing and feedback loops refined the formulation over a few rounds, yielding a composition suitable for animal studies.
After vacuum‑drying, the heat‑resistant LNPs stored at high temperature still elicited immune responses in mice comparable to those from the original Moderna formulation. The same formulation was also cast into solid microneedle patches, delivering comparable immunity. The algorithm successfully adapted to a Pfizer‑type LNP by adjusting excipient ratios.
This approach can rapidly generate thermally stable formulations for any mRNA payload, expanding the reach of vaccines and other advanced delivery platforms.
Review