(M1130-01-05) Toward Enhanced Stability of mRNA Lipid Nanoparticles

Purpose: Messenger RNA (mRNA) is an evolving class of nucleic acid-based therapeutics. In particular, mRNA vaccines have the advantages of modularity and high manufacturing capabilities. Therefore, it may be ideal platform for protection against mutagenic viruses and the eradication of pandemics. Unfortunately, the FDA-approved COVID-19 mRNA vaccines require freezing temperatures as low as -90°C for storage and distribution. Storage and distribution requirements of mRNA vaccines are a major challenge toward mass immunization particularly in low-income countries. Ultimately, enhanced lipid nanoparticle compositions and formulation strategies are required to address the cold reality of mRNA vaccines and other mRNA lipid nanoparticle-based therapeutics. Herein, we employed Taguchi fractional factorial design of experiments (DoE) to investigate the effect of lipid molar ratios on the stability of mRNA lipid nanoparticles (LNPs). Additionally, we employed Thin Film Freeze Drying (TFFD) technology to engineer dry powder compositions for mRNA LNPs that can potentially address the ultracold temperature requirements for storge and distribution.

Methods: Taguchi L16 (4⁴) fractional factorial DoE was employed to formulate 16 LNP formulations comprising DLin-MC3-DMA (molar ratio 30, 40, 50 or 60), DSPC (molar ratio 5, 10, 15 or 20), cholesterol (18.5, 28.5, 38.5, or 48.5), and PEG₂₀₀₀-DSPE (molar ratio 0.5, 1,1.5, or 2). Ploy(A) (MW 700-3,500) was loaded into the LNPs as a model mRNA at a nitrogen: phosphate ratio of 3:1. Poly(A) LNPs were prepared by nanoprecipitation technique via rapidly mixing the aqueous and organic phases at a ratio of 3:1 v/v by pipetting. Poly(A) LNPs were then dialyzed overnight against 250× sample volume of RNase-free water. The average particle size of Poly(A) LNPs was studied using dynamic light scattering and the Poly(A) encapsulation efficiency (EE) was determined by the Quant-iT Ribogreen RNA assay. The Poly(A) LNP stability was studied after storage at 4°C for 2 days. The effect of freezing rate on the average particle size of Poly(A) LNP was studied by slowly freezing the Poly(A) LNP at a rate of 1°C/min or fast freezing by dropping the liquid formulations on a cryogenically cooled drum surface (-80°C). Data were analyzed using LASSO (least absolute shrinkage and selection operator) regression. A dry powder of formulation F6 was developed using TFFD. Trehalose was employed as a cryoprotectant at a final concentration of 2.5, 5, 7.5, 10 or 15 %w/v. The frozen thin films were dried at various primary drying times (i.e., 8, 10, 20 or 40 h) at a shelf temperature of -40°C, while the secondary drying time was 1 h at +4°C. The chamber pressure was kept at 80 mTorr. The water content of the powders was determined using a coulometric Karl Fischer titrator.

Results: The minimum for initial particle size and maximum for initial EE was found at the highest cholesterol molar ratios compared to other lipids. High ratios of DLin-MC3-DMA were detrimental to the particle stability after 2 days of storage at +4°C. The interaction between DLin-MC3-DMA and PEG₂₀₀₀-DSPE helps maintain the particle size during fast freezing. The water content of dry powder prepared using TFFD affected the average particle size and EE. For optimal stability, the moisture content in the powder should be kept between 1 and 3% w/w. Excessive drying and insufficient drying can be detrimental to particle size. After optimizing the lyophilization cycle and hence the residual moisture content, dry powder formulations of poly(A) LNPs were successfully formulated at a trehalose concentration as low as 2.5% w/v.

Conclusion: The lipid nanoparticle composition can be tailored to develop LNPs suited for stability under certain conditions (e.g., in liquid form at +4°C or against freezing and thawing). Dry powder formulation can help stabilize the mRNA LNPs; however, the drying time and residual water content should be optimized. A water content between 1 and 3% is required for optimal stability.

References: 1. AboulFotouh, K., Z. Cui, and R.O. Williams, 3rd, Next-Generation COVID-19 Vaccines Should Take Efficiency of Distribution into Consideration. AAPS PharmSciTech, 2021. 22(3): p. 126.
2. Schoenmaker, L., et al., mRNA-lipid nanoparticle COVID-19 vaccines: Structure and stability. Int J Pharm, 2021. 601: p. 120586.
3. Pardi, N., et al., mRNA vaccines - a new era in vaccinology. Nat Rev Drug Discov, 2018. 17(4): p. 261-279.
4. Weng, Y., et al., The challenge and prospect of mRNA therapeutics landscape. Biotechnol Adv, 2020. 40: p. 107534.

Acknowledgements: This work was supported by Sponsored Research Agreements and Technology Validation Agreements from TFF Pharmaceuticals Inc. (to ROW and ZC). KA is supported in part by a fellowship (GM 1105) from the Egyptian Ministry of Higher Education. Conflicts of interest statement: ZC and ROW report financial support by TFF Pharmaceuticals, Inc. ZC reports a relationship with TFF Pharmaceuticals, Inc. that includes: equity or stocks and funding. ROW reports a relationship with TFF Pharmaceuticals, Inc. that includes: consulting or advisory, equity or stocks, and funding. ZC and ROW have a patent “Dry solid aluminum adjuvant-containing vaccines and related methods thereof” pending to TFF Pharmaceuticals, Inc. ZC, ROW, KA and HX have a patent “Dry liposome adjuvant-containing vaccines and related methods thereof” pending to UT Austin. ZC, ROW and KA have two patents “Dry liposome formulations and related methods thereof” and “Dry powders of biologics and nucleic acid-based products for intranasal administration: compositions and method of preparing” pending to UT Austin.

Fig. 1. The perspective plots of the effect of lipid molar ratios on (A) initial particle and (B) initial EE. (C-D) The perspective plots of the effect of lipid molar ratios on particle size and EE after storage of liquid formulations for 2 days at +4°C. (E-F) The perspective plots of the effect of lipid molar ratios on average particle size and EE after fast freezing at -80°C and thawing at +4°C.

Fig. 2. Effect of powder moisture content on (A) a average particle size of F6, (B) EE of F6, (C) average particle size of F9 and (D) EE of F9.

Fig. 3. Characterization of poly(A) LNPs. (A) Mean particle size of poly(A) LNPs containing various concentrations of trehalose before TFFD and after TFFD and reconstitution and (B) EE of poly(A) in the LNPs before TFFD and after TFFD and reconstitution. Formulations containing different trehalose concentrations showed similar poly(A) EE in the LNPs before TFFD. Data are mean ± S.D. (n =3). ns: no significant difference between the means of different groups, *p≤0.05, **p≤0.01, ****p≤0.0001. (C) SAXS scattering profile of F6 in 5% w/v trehalose before (LIQUID) and (D) after TFFD. (E-F) Electron density reconstructions from experimental solution scattering data for F6 in 5% w/v trehalose (E) before, and (F) after TFFD. Electron density are shown in surfaces with 1, 2, 5, 10 σ from the average electron density of the particle.