Key findings include:
- The metabolism of ionizable lipids is primarily limited by the LNP disassembly rate rather than the hydrolyzation of the lipids.
- The RNA release ability from endosomes was quantitatively assessed, predicting RNA release probability.
- The biodegradability of ionizable lipids, as estimated by QM methods, aligned with PBPK model results.
The integrated multi-source data and multi-level modeling provided new insights into the biodegradability, safety, and RNA release mechanisms of LNPs, aiding future design and research of RNA-LNP formulations.
Lipid nanoparticle (LNP) delivery systems have revolutionized nucleic acid therapies, leading to approvals of siRNA drugs like Onpattro® and mRNA vaccines like Comirnaty® and Spikevax®. LNPs protect nucleic acids from degradation and facilitate their delivery into target cells. A typical LNP consists of ionizable lipids, cholesterol, phospholipids, and polyethylene glycol (PEG) lipids, each playing a specific role in the delivery process. Ionizable lipids, crucial for encapsulating RNA and facilitating cellular uptake and endosomal escape, are a defining feature of LNPs. Understanding the pharmacokinetic (PK) properties of LNPs is vital, as PK issues are a significant cause of drug development failures.
PBPK and QM Modeling Approaches
Physiologically-Based Pharmacokinetic (PBPK) Modeling: PBPK models simulate PK based on physiological and pharmaceutical parameters and are recommended for drug development. For LNP formulations, these models can simulate RNA-LNP delivery in various organisms (e.g., Hela cells, rats, mice, humans) and investigate the PK of ionizable lipids and RNA simultaneously.
Quantum Mechanics (QM) Modeling: QM modeling, particularly useful for simulating drug metabolism, examines the interaction of molecules at a microscopic level. This study used QM to model the metabolism of ionizable lipids, providing insights into their biodegradability and potential toxicity.
Key Findings
-
In Vivo PBPK Modeling:
- Model Structure: An optimal model structure was determined through iterative fitting to rat PK data, considering factors like LNP permeability, cellular uptake, disassembly, and metabolism rates.
- Ionizable Lipids: The model showed differences in the PK behaviors of MC3, SM-102, and Lipid 5, highlighting the importance of disassembly rates over metabolism rates in determining biodegradability.
- Size and Dose: LNP size significantly impacts PK, with smaller LNPs showing higher uptake rates but differing in gene knockdown efficiency. Human PK data of patisiran (Onpattro®) confirmed the model's applicability across different doses.
-
Cellular PBPK Modeling:
- Physiological Parameters: Parameters for LNP transportation within cells were identified, providing a detailed mechanistic understanding of LNP uptake, disassembly, and RNA release.
- Comparative Analysis: The study compared the RNA release efficiency of LNPs containing different ionizable lipids (C12-200, MC3, L319), revealing differences in release rates and probabilities of RNA undergoing release events.
-
QM Modeling:
- Metabolism Simulation: QM modeling estimated the energy changes during lipid hydrolysis, indicating the relative biodegradability of MC3, SM-102, and Lipid 5, aligning with PBPK model findings.
Conclusion and Future Perspectives
The integration of PBPK and QM modeling provides a comprehensive framework for understanding and optimizing LNP formulations. Despite some limitations (e.g., simplified model structures, assumptions regarding receptor concentrations), this multi-level modeling approach advances the research of nanomedicines by bridging macroscopic PK properties with microscopic mechanistic details. Future work should focus on refining models to incorporate more detailed physiological processes and exploring the activation energy of lipid hydrolysis for more precise kinetic prediction.
