Messenger RNA vaccine platforms from molecular optimization to infectious disease and cancer immunotherapy: a narrative review
Year: 2026
Type: review
Abstract: Background and Objective: Messenger RNA (mRNA) vaccines now encompass licensed respiratory vaccines, a self-amplifying ribonucleic acid (RNA) product, and individualized cancer immunotherapies in late-stage development. The expansion of the field has made platform-level comparisons increasingly important because efficacy, tolerability, manufacturability, and tissue targeting are influenced by the RNA architecture as well as by the delivery system. This narrative review examines molecular engineering, delivery, emerging RNA architectures, infectious-disease translation, and cancer immunotherapy, with emphasis on developments reported through 2026. Methods: Targeted literature searches were conducted in PubMed through 31 August 2026, complemented by reference chaining and targeted searches of official regulatory sources. English-language mechanistic studies, clinical trials, regulatory documents, and focused reviews were prioritized. Particular attention was given to evidence published from 2023 to 2026 on self-amplifying RNA (saRNA), circular RNA (circRNA), molecular optimization, and individualized neoantigen vaccines. Key Content and Findings: Nucleoside-modified non-replicating mRNA formulated in lipid nanoparticles remains the most clinically validated architecture. Molecular optimization now extends beyond nucleoside substitution to cap chemistry, untranslated-region selection, coding-sequence and RNA-structure design, poly(A) control, and removal of double-stranded RNA (dsRNA) impurities. saRNA has reached regulatory authorization and offers dose-sparing potential, whereas circRNA remains predominantly preclinical and still faces unresolved manufacturing and quality-control questions. Infectious-disease programs have established reusable capabilities in antigen design, lipid nanoparticle formulation, manufacturing, and regulatory comparability. In oncology, intismeran autogene plus pembrolizumab has produced durable recurrence-free survival (RFS) benefit in resected melanoma, while pancreatic-cancer studies of autogene cevumeran show that vaccine-induced T cell responses vary substantially between patients. Tumor clonality, neoantigen quality, antigen presentation, the immune microenvironment, treatment setting, and manufacturing time are therefore as important as the number of encoded mutations. Artificial intelligence (AI) can improve neoantigen ranking but does not remove the need for biological validation. Conclusions: No RNA architecture is uniformly superior across indications. Further clinical progress will depend on matching transcript design and delivery to the intended immune response, improving predictive antigen selection and patient stratification, and transferring validated manufacturing and regulatory capabilities from infectious-disease vaccines to individualized oncology without assuming that platform maturity alone is sufficient for antitumor efficacy.
Source: Translational Cancer Research
Authors: Qian Wang, Zhangzhang Sun, Heyong Liu
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: diamond
OpenAlex ID: W7224112381