Self-amplifying RNA (saRNA) is a promising platform for the production of vaccines, anti-tumor therapeutics, and gene therapy solutions. One of the advantages of the saRNA platform is the ability to use small doses of the therapeutic while maintaining prolonged expression of the target protein. However, the presence of auxiliary sequences encoding non-structural alphavirus proteins, which facilitate the replication of saRNA in cells, necessitates a thorough assessment of the biosafety of this platform. In our review, we focus on saRNA functions in the context of its interaction with the innate immune system. Firstly, an analysis is conducted of the side effects of candidate saRNA therapeutics, as observed in preclinical and clinical trials. Then, the mechanisms underlying the function of saRNA products derived from various alphavirus genomes in cell systems are discussed, as well as the reasons for their reactogenicity. The key approaches to optimizing the saRNA platform, which are aimed at reducing the activation of the innate immune response and cytopathic effects, are described. To summarize, this review enables us to systematize our knowledge on the advantages and disadvantages of saRNA, as well as potential approaches to improving this platform in order to develop more effective and safer therapeutics.

Background: Lipid nanoparticles (LNPs) can activate innate immunity, but the factors shaping repeat-dose reactogenicity remain unclear. We compared systemic, local, and organ-specific reactions after intramuscular administrations of empty LNPs containing ionizable lipids used in mRNA vaccines. Methods: Female BALB/c mice received two intramuscular injections, 2 weeks apart, of LNP-4 (four components, ALC-0315 as the sole ionizable lipid; 59 nm) or LNP-5 (five components, equimolar ALC-0315/SM-102 at the same total ionizable lipid content; 86 nm). Body weight was monitored for 15 days. Blood biochemical parameters, injection-site histology, and morphological and molecular markers in the hypothalamus, liver, lungs, and spleen were assessed 24 h after the second injection by histology, immunohistochemistry (IHC), immunofluorescence (IF), Western blotting, and quantitative PCR (qPCR). Results: Both LNPs were followed by transient body-weight loss, moderate biochemical changes, and local reactogenicity. The strongest organ-specific responses occurred in the liver and hypothalamus. Both LNPs induced hepatic sinusoidal dilation and leukocyte infiltration, but LNP-4 was associated with more myeloid cell numbers and higher inflammatory marker expression. In the hypothalamus, LNP-4 produced broader changes in inflammatory and myeloid marker expression than LNP-5. No marked structural damage was observed in the examined organs. Conclusions: Repeated LNP administration was followed by systemic and local reactogenicity, and organ-specific changes differed between the formulations, possibly reflecting lipid composition, particle size, and lipid dose. Carrier composition should be considered in the design and safety evaluation of LNP-based vaccine delivery systems.