Abstract

Exosomes, a category of extracellular vesicles (EVs), are being engineered as biocompatible nanocarriers for gene and genome-editing agents, such as messenger RNA, small interfering RNA/antisense oligonucleotides, plasmid DNA, and CRISPR/Cas (clustered regularly interspaced short palindromic repeats / CRISPR-associated) systems. Their intrinsic membrane composition can safeguard sensitive cargos, diminish immunogenicity, and enable repeated administration; nevertheless, translation is hindered by inconsistent isolation and characterization, variable cargo loading, non-specific biodistribution, and challenges in manufacturing scale-up. This review consolidates recent advancements in the engineering of exosomes for therapeutic gene delivery and introduces an integrated AI-enabled design framework. This framework utilizes AI-assisted whole-exome sequencing variant prioritization to guide target selection, modality determination, and exosome engineering choices, including loading strategies, surface targeting, and mechanism-linked potency assessments. We detail novel exosome-mediated CRISPR delivery methods, critically evaluate safety concerns and potential failure modes, and delineate a pragmatic translational roadmap consistent with expectations for EV standardization. The integration of patient genetics with vector design and manufacturing quality characteristics through AI-exosome convergence provides a pathway to more systematic, reproducible, and clinically scalable gene treatments.

Key words: Exosome, Gene Delivery, Whole Exome Sequencing (WES), CRISPR-Cas9, Genomics, Artificial Intelligence, Targeted Nanomedicine.