Potentiate mRNA Lipid Nanoparticle-mediated Anticancer Immunity by Optimizing Th1 and Th2 Immune Activation Profile

Introduction: Lipid nanoparticles (LNPs) have been designed as immunostimulatory delivery tools for antigen-encoding mRNA in cancer immunotherapy. Primary efforts have focused on engineering LNPs to promote transfection and maturation of antigen-presenting cells (APCs) and modulate adjuvant activity to elicit strong cytolytic CD8+ T cell responses for improved anti-tumor efficacy. However, the roles of LNP formulation compositions, particularly the charge of the helper lipid and the relative ratios of the LNP components, in polarizing specific subtypes of immune responses have remained poorly understood. In this study, we explore the potential of optimizing Type-1 T helper (Th1) and Type-2 T helper (Th2) immune activation profiles by tailoring LNP composition for modulation of transfection efficiency in APCs vs. non-APCs, thereby potentiating mRNA lipid nanoparticle-mediated antitumor immunity. Here, we employ a multi-step screening method to optimize the type of helper lipid and component ratios in an mRNA LNP library to efficiently deliver antigen-encoding mRNA to distinct cell populations and modulate the immune activation profile. By first screening for APC maturation and antigen presentation in vitro followed by assessing immune activation and tumor growth suppression in vivo, we investigate the orchestration of different immune subtype responses against tumor antigens generated by distinct LNP compositions. This approach allows us to examine the relationships among LNP composition, transfection efficiency in APCs vs. non-APCs, immune subtype activation profile, and the resulting anti-tumor efficacy of LNP-based mRNA cancer vaccines.

Materials and

Methods: An LNP library was created by using DLin-MC3-DMA as the ionizable lipid, DMG-PEG2000 as the PEGylated lipid, cholesterol, and six helper phospholipids (DOTAP, DDAB, DOPE, DSPC, 14PA, and 18PG) that were previously used in FDA-approved or experimental LNP formulations, representing a diverse array of phospholipid charges and structures. By varying the relative weight ratios of the lipid components and the molar ratio of chargeable groups in ionizable lipid to phosphate groups in the nucleic acid cargo (N/P ratio), a series of 1,080 LNP formulations were generated and screened for transfection efficiency in dendritic cells (DCs) in vitro using luciferase reporter mRNA. Antigen presentation and co-stimulatory molecule expression by DCs following antigen delivery were assessed in vitro using LNP formulations loaded with ovalbumin (OVA) mRNA. LNPs with the highest DC transfection, antigen presentation, and maturation were further examined for the transgene expression levels and immune activation profiles following subcutaneous (s.c.) injection. The prophylactic and therapeutic efficacy of the lead mRNA LNP formulations as cancer vaccines were evaluated in B16F10, MC38, and EG7 tumor models using OVA as the model antigen, as well as syngeneic tumor models using mRNAs encoding Trp2 and Gp100 tumor antigens.

Results, Conclusions, and Discussions: Based on the in vitro library screening, seven LNP formulations achieved the highest transfection of freshly isolated bone marrow-derived DCs (BMDCs) with >70% cells transfected, and three of them (C10, D6, and F5) demonstrated elevated antigen presentation (Fig. 1A). In vivo assessments through s.c. injections indicated that these top three formulations, C10, D6, and F5, led to 4.0-, 2.7-, and 3.8-fold higher numbers of SIINFEKL-H-2Kb+CD11c+ cells in dLNs than the PBS group, respectively (Fig. 1B). Interestingly, immune activation evaluations showed that, while a significantly higher number of antigen-specific Th1 (CD4+IFN-γ+T-bet+) cells was observed for all three formulations, C10 showed a much stronger Th2 response (CD4+IL-4+GATA3+) along with a markedly higher OVA-specific IgG titer (Fig. 1C–E). Given the potent antigen-specific immune responses, these three LNP formulations were examined for their prophylactic and therapeutic efficacies against cancer. In a B16-OVA melanoma prophylactic model, all three LNP formulations showed substantially stronger tumor growth inhibition with prolonged overall survival times than the PBS group; more interestingly, C10 formulation, which elicited both strong Th1 and Th2 responses, yielded a markedly improved protection effect with 18 out of 30 mice remaining tumor-free beyond 100 days (Fig. 1F, G). In addition, four out of nine mice remained tumor-free after being rechallenged with B16F10-OVA cells, suggesting the sustained protection conferred by C10 mOVA LNPs (Fig. 1H). Notably, a significant tumor growth delay was observed in mice challenged with B16F10 cells (Fig. 1I), likely due to epitope spreading. Additionally, C10 LNPs displayed potent tumor suppression in B16F10-OVA, MC38-OVA, and EG7-OVA treatment models. Strong anti-tumor effects were also observed in syngeneic tumor models using mRNAs encoding Trp2 and Gp100 tumor antigens, showing prolonged median survival times of 23 and 23.5 days for C10-mTrp2 LNPs and C10-mGp100 LNPs, respectively. In conclusion, this LNP screening platform allows for the identification of the best-performing mRNA LNPs for APC-specific gene expression and Th1/Th2 skewed immune modulation. The LNP formulation that is capable of eliciting both potent Th1 and Th2 responses yields the most potent antitumor efficacy and long-term memory response compared to formulations with only Th1-skewed responses, offering new opportunities for RNA-based vaccine optimization.

Acknowledgements (Optional):