MIT News breed · Science

AI-Assisted Formulation Enhances Heat Resistance of RNA Vaccines

Researchers have developed a new formulation for lipid nanoparticles (LNPs) used in RNA vaccines, significantly improving their heat resistance and potentially simplifying global distribution.

Using an AI algorithm, MIT scientists modified LNP formulations for mRNA vaccines, making them stable at room temperature for up to a year or at nearly 100°F for two months. This addresses the ultracold storage limitation of current RNA vaccines.

In mice, these heat-resistant RNA COVID-19 vaccines showed immune responses comparable to a Moderna-like vaccine. The AI accelerated development by predicting optimal formulations, reducing experiments.

RNA's fragility requires LNP stabilization, but cold-chain needs hinder distribution. The new formulation could enable wider access and novel delivery methods like microneedle patches.

The team focused on enhancing FDA-approved LNP formulations. After initial challenges with known excipients, they used machine learning to predict effective excipient ratios.

The AI analyzed nearly 50 FDA-approved excipients for RNA stabilization within LNPs. Promising excipients underwent AI-driven predictions and tests, yielding a formulation for animal studies in weeks.

Vaccines in the new formulation, stored at elevated temperatures, induced robust immune responses in mice. The formulation also enabled microneedle patches with comparable immune responses.

This approach applies to various LNP formulations, including Pfizer's, and can be adapted for different mRNA payloads, benefiting other LNP-based therapeutics.

AI-samenvatting op basis van de bron.