Engineered multi-domain lipid nanoparticles for targeted delivery

Abstract

Engineered lipid nanoparticles (LNPs) represent a breakthrough in targeted drug delivery, enabling precise spatiotemporal control essential to treat complex diseases such as cancer and genetic disorders. However, the complexity of the delivery process—spanning diverse targeting strategies and biological barriers—poses significant challenges to optimizing their design. To address these, this review introduces a multi-domain framework that dissects LNPs into four domains: structure, surface, payload, and environment. Engineering challenges, functional mechanisms, and characterization strategies are analyzed across each domain, along with a discussion of advantages, limitations, and in vivo fate (e.g., biodistribution and clearance). The framework also facilitates comparisons with natural exosomes and exploration of alternative administration routes, such as intranasal and intraocular delivery. We highlight current characterization techniques, such as cryo-TEM and multiscale molecular dynamics simulations, as well as the recently emerging artificial intelligence (AI) applications—ranging from LNP structure screening to the prospective use of generative models for de novo design beyond traditional experimental and simulation paradigms. Finally, we examine how engineered LNPs integrate active, passive, endogenous, and stimuli-responsive targeting mechanisms to achieve programmable delivery, potentially surpassing biological sophistication in therapeutic performance.