Explore recent mRNA, saRNA, circular RNA, and LNP research on delivery, therapeutic translation, safety, analytical characterization, and manufacturing.
Selected for substantive contributions to delivery, translation, analytics, manufacturing, and safety. Original publication dates and evidence levels are retained. This library covers the latest 120 days; individual published guides remain available.
Meng Tian; Leiming Wang; Zhengwei Liu; Dinglingge Cao; Changyue Yu; Jeffrey L. Bennett; Eliza Bliss-Moreau; Chang Wang; Diana D. Kang; Yonger Xue; Yichen Zhong; Siyu Wang; Haoyuan Li; Xucheng Hou; Binbin Deng; Yizhou Dong Nature Nanotechnology · DOI 10.1038/s41565-026-02326-7
Researchers used the properties of more than 100 central nervous system drugs to design brain-directed ionizable lipids. HX1 nanoparticles delivered RNA and enabled gene editing in mouse brains, delivered reporter mRNA in a nonhuman primate, and reduced radiation-related brain injury in mice carrying Il10 mRNA. These are preclinical findings.
Why it is included: Brain delivery could expand RNA medicine beyond established targets, but repeat-dose safety and human translation still need validation.
Wang G; Guo Y; Guo K; Zhang C; Zhou Y; Xia J; Jiang H; Liu J; Ren J; Mohamed AO; Tang K; Chen X; Sun S; Xing Y; Yang Y; Shen M; Wang R; Huang Y; Li X; Lin A Molecular therapy : the journal of the American Society of Gene Therapy · DOI 10.1016/j.ymthe.2026.10.021
An engineered nanoparticle delivered mRNAs encoding IL-1 receptor antagonist and an angiopoietin-like 3 derivative through intra-articular injection. The combination addressed inflammation and cartilage degeneration in several mouse arthritis models, with transcriptomic analyses supporting complementary biological effects. These findings broaden local mRNA therapy beyond vaccination, but durable structural benefit, repeated-dose safety, and effectiveness in human joints remain unproven.
Why it is included: Explores local protein replacement for two joint-disease processes; mouse models cannot establish durable cartilage regeneration or disease modification in patients.
Researchers combined a 520-member degradable lipid library with interpretable machine learning and a 10,920-member virtual design space. Independent mouse experiments identified PA97MD101 as a lead for hepatic mRNA expression, Cre-mediated recombination, and Cas9-based TTR suppression. The study links screening predictions with functional animal validation, without establishing clinical safety or efficacy.
Why it is included: Combining interpretable models with independent animal testing may improve lipid selection beyond cell-based screens and guide more efficient formulation discovery.
Chun K; Choi SY; An K; Jo DH; Lee H Journal of controlled release : official journal of the Controlled Release Society · DOI 10.1016/j.jconrel.2026.115453
A C12-200-based formulation was optimized for movement and cellular delivery in the protein-poor vitreous environment. It co-delivered RB1 mRNA and triptolide, improving antiproliferative activity and suppressing tumors in an orthotopic retinoblastoma model after intravitreal administration. Retinal evaluations supported tolerability during the study period. The combination remains preclinical and requires longer ocular safety assessment and pediatric translation studies.
Why it is included: Demonstrates tissue-environment-specific engineering for ocular combination therapy; animal tumor suppression and short-term retinal assessments do not establish pediatric clinical safety.
Jinbi Tian; Khanh T. M. Tran; Brett H. Pogostin; Olivia Sheridan; Sevinj Mursalova; Amy H. Lee; Shuai Liu; Jaya Hamkins; Daniel Antov; Alana L. Power; Zane S. Dash; Dongsoo Yun; Mina Konaković Luković; Robert S. Langer; Ana Jaklenec Nature Biotechnology · DOI 10.1038/s41587-026-03331-w
The AGENT platform combines Bayesian optimization with high-throughput experiments to identify thermostable mRNA-LNP formulations through six iterations in approximately one month. Solid-state model formulations based on SM-102 and ALC-0315 retained biological activity after more than two months at 37°C. Animal immunogenicity supported further development, but these formulation-specific findings do not eliminate cold-chain requirements for mRNA products generally.
Why it is included: Illustrates efficient formulation optimization for thermostability; real products require long-term potency, transport, reconstitution, and safety data under their intended conditions.
Yuxin He; Lin Wang; Abigael P. Williams; Trisha I. Valerio; Yuanhong Sun; James Bowman; Samantha D. Ricketts; Savannah Hammond; Aidan Sison; Jezan Alexandre; Zachary Niastor; Wei R. Chen; Dixy E. Green; Kar-Ming Fung; Paul L. DeAngelis; Stefan Wilhelm ACS nano · DOI 10.1021/acsnano.6c11865
Heparosan surface modification was compared with conventional PEG-based nanoparticles across several ionizable lipid formulations. Modified particles improved expression in immune-cell models and showed reduced association with late endosomes and lysosomes. Mouse experiments assessed repeated expression and vaccine antibody responses. The findings support an alternative surface-engineering approach, while long-term immunogenicity, biodistribution, and manufacturing consistency require additional study.
Why it is included: Expands surface-engineering options for immune delivery; trafficking correlations and short-term mouse tolerability do not establish long-term human safety or universal PEG replacement.
Nageen Sardar; Shoaib Manzoor Human immunology · DOI 10.1016/j.humimm.2026.112083
This review compares self-amplifying and circular RNA vaccine architectures, expression kinetics, delivery approaches, and immune responses. It discusses the different reasons these platforms may extend antigen production or reduce dose requirements, alongside manufacturing, purity, reactogenicity, and regulatory challenges. It is a synthesis of heterogeneous preclinical and early clinical evidence, not a new head-to-head trial establishing superiority over conventional mRNA vaccines.
Why it is included: Useful for comparing next-generation RNA platforms; heterogeneous studies and limited clinical data prevent broad claims of superior durability or safety.
Ao Hoi Tong; Haichao Zhu; Chenming Zou; Bingshu Liang; Xueer Wu; Shengrong Guo International journal of biological macromolecules · DOI 10.1016/j.ijbiomac.2026.154367
An injectable hydrogel co-delivered hyaluronidase and light-responsive nanoparticles carrying IL-12 and GM-CSF mRNAs. The design combined stromal remodeling, local retention, and photoactivated intracellular delivery. In a postoperative mouse breast-cancer model, treatment reduced recurrence and altered local and systemic immune responses. The multiple-component system remains experimental, with light access, long-term safety, and clinical feasibility requiring further work.
Why it is included: Integrates extracellular and intracellular delivery strategies after surgery; a mouse recurrence model and short-term safety checks cannot establish clinical recurrence prevention.
Vita Golubovskaya; Liang Hu; Shiming Zhang; John Sienkiewicz; Hua Zhou; Robert Berahovich; Jinying Sun; Le Li; Walter Bodmer; Lijun Wu Proceedings of the National Academy of Sciences of the United States of America · DOI 10.1073/pnas.2616059123
Investigators combined mRNAs encoding an EpCAM-directed T-cell engager, IL-12, and GM-CSF through intratumoral nanoparticle administration. In partially humanized mouse xenograft experiments, the combination altered immune activity in both injected and uninjected tumors and inhibited tumor growth. The study supports localized production of several immune proteins, while the model cannot establish safety or effectiveness across human cancers.
Why it is included: Supports local multicomponent mRNA immunotherapy and distant-tumor effects; partially humanized mouse models provide limited predictions of human systemic toxicity.
The branched ionizable lipid FL0445 supported delivery of linear mRNA and capped circular RNA through a cholesterol-dependent cellular-entry pathway. In obese mice, GLP-1-encoding capped circular RNA produced sustained expression and improved glucose tolerance relative to tested formulations. Inflammatory transcriptional signals were also assessed. The results provide a preclinical delivery proof of concept, not evidence of clinical metabolic efficacy.
Why it is included: Links circular-RNA architecture with carrier selection and duration; mouse metabolic effects do not establish human dosing, safety, or therapeutic benefit.
Researchers used two-armed PEG conjugated to oligocations to transiently coat liver sinusoidal endothelium before mRNA-LNP exposure. The approach modifies the host delivery barrier rather than the LNP itself, aiming to reduce hepatic entry and off-target expression. Mouse experiments examined reversible coating and clearance. Translation would require evaluating the combined regimen, vascular effects, and repeat-dose safety.
Why it is included: Explores host-barrier modulation as a complement to particle engineering; clinical safety cannot be inferred from transient mouse liver shielding.
Siyu Wang; Yichen Zhong; Chang Wang; Meng Tian; Jeffrey L. Bennett; Eliza Bliss-Moreau; Yonger Xue; Changyue Yu; Xucheng Hou; Ya Ying Zheng; Haoyuan Li; Zhengwei Liu; Dinglingge Cao; Diana D. Kang; Binbin Deng; Yizhou Dong Proceedings of the National Academy of Sciences · DOI 10.1073/pnas.2538147123
The CA2d ionizable-LNP platform delivered mRNA to brain cells after intravenous administration in mice and nonhuman primates. Reported target populations included neurons, microglia, and astrocytes. In a rat ischemic-stroke model, combined thrombolytic and neuroprotective mRNA cargos reduced injury and improved functional measures. The work expands preclinical CNS delivery evidence without demonstrating human brain delivery or treatment efficacy.
Why it is included: Adds nonhuman-primate evidence for systemic CNS RNA delivery; species differences and neurological safety remain major translational questions.
