Scientists have found a new way to stabilise ribonucleic acid (RNA) vaccines during heat period.
The new approach addresses one of the biggest challenges with RNA vaccines, like those used against COVID-19.
Due to their fragility, such vaccines must be stored at ultracold temperatures, making distribution difficult in regions without specialised freezers.
Now MIT engineers have developed a way to stabilise the lipid nanoparticles (LNPs) that carry RNA, potentially allowing vaccines to remain effective at room temperature for up to a year.
RNA is notoriously unstable, so scientists encase it in lipid nanoparticles to protect it and help it enter cells. Even then, current RNA-LNP vaccines require storage at –20 to –80 °C.
The MIT team, led by Ana Jaklenec and Robert Langer, used an AI algorithm to redesign the nanoparticle formulation, making it more heat-resistant.
The new formulations remained stable at room temperature for one year and at 98 °F (37 °C) for two months. When tested in mice, the vaccines generated immune responses just as strong as those from Moderna-like RNA vaccines.
“The real beauty of this algorithm is that we can use it with small data sets. It’s really hard to run thousands of experiments, so this algorithm allows us to more easily achieve formulations with features that we want, in this case, stability,” Jaklenec explained.
The team worked with MIT’s Computer Science and Artificial Intelligence Lab (CSAIL) to develop a machine-learning model that could predict optimal excipient combinations. Excipient molecules, such as sugars, salts, or polymers, are added to stabilise LNPs.
Rather than using trial and error with hundreds of excipients to find the best formulation, the algorithm used data from approximately fifty FDA-approved excipients to rapidly identify potential formulation candidates.
This method allowed the research team to discover a promising candidate for further development in a matter of weeks, as opposed to what would have taken them months if they had not received any assistance from an AI-based system.
“It was surprising to see how quickly the algorithm converged on a stable formulation, getting there in just a handful of iterations,” said Mina Konaković Luković, assistant professor at CSAIL.
The implications of stable RNA formulations extend far beyond pandemic vaccines. They could expand global distribution, particularly in regions lacking cold-chain infrastructure. They may enable new delivery methods, such as microneedle patches that dissolve into the skin.
The ability to use RNA stability as a foundation for the next generation of medical treatments would be enabled by its potential use in providing therapeutic options and developing new delivery systems for drugs, such as controlled-release particles. Together, these advances will allow us to establish RNA stability as a fundamental component of the next generation of medicine.
Graduate student Jinbi Tian noted: “Our approach broadens the application of not only mRNA vaccines, but also therapeutics or advanced drug-delivery platforms like controlled-release particles or microneedle patches.”
The team also showed that their algorithm could stabilise formulations similar to those used by Pfizer’s Covid-19 vaccine, suggesting broad applicability. Once a heat-resistant LNP formulation is developed, it can be adapted to deliver any type of mRNA payload.
This work, published in Nature Biotechnology, represents a major step toward making RNA vaccines more practical, accessible, and versatile. It’s a striking example of how AI and biology can work hand in hand.
By letting algorithms guide formulation design, researchers can leapfrog years of trial and error and bring life-saving technologies closer to the people who need them most.
