LNP meets high-throughput Cryo-EM: Reliably quantifying LNP morphology
The never-ending quest to develop and establish more efficient drug delivery systems has led to significant innovations in the rapidly advancing field of nanoparticle drug delivery technology. At its forefront, Lipid Nanoparticles (LNPs) have emerged as a cornerstone for delivering therapeutic agents in the human body. This article highlights how ATEM’s high-throughput cryo-EM technology for the first time enables researchers to extract and characterize statistically relevant, accurate and precise quantitative insights into LNP morphology and size. The reliable quantification of key LNP quality attributes offers significantly more advanced insights to rationalize development decisions during formulation, precisely monitor batch-to-batch consistently and evaluate the morphological effects on LNPs during stability testing.
The Critical Role of Morphology in LNP Efficacy
The “Structure-to-Function” relationship represents a fundamental principle in nature. Not only applicable in physic, where the exact composition and appearance of sub-atomic particles critically determine the properties of each element, or in chemistry, where the exact interconnection and structure of individual atoms define the fundamental properties of each and every molecule we know, but also in nanoparticle technology: the Structure, hence, morphology of every particle is inseparably linked to its intended (or unintended) function.
The morphology of LNP particles can be diverse and, unsurprisingly, turns out to be of paramount importance for the potency, stability, and successful delivery of every formulated product. Ongoing research from academia and industry converges towards the result that distinct morphological features, such as particle size, shape, and intra-particle morphology (morphology class), successfully forecast the potency and stability of the LNP formulation in question. 1. 2. 3. Being able to quantitatively analyze and compare LNP morphology thus represents a remarkable opportunity to rationally optimize LNP formulations and predict their stability based on the precise monitoring of key morphological features.
FDA Recommendations & Academic Insights
The importance of characterizing morphology in LNP and nanoparticle-based drugs extends beyond the need to generate more insightful analytical data to optimize or enable drug candidate potency and stability. As the Structure-to-Function relationship is particularly critical for Nanoparticle formulations, regulatory authorities concerned about the efficacy and quality of drug products, such as the U.S. Food and Drug Administration (FDA), recommend the analysis of key morphological features of every nanoparticle drug (candidate). These include, but are not limited to: accurate particle size distribution, general shape, morphology, and aspect ratio. In effect, precise characterization of these key quality attributes is pivotal in meeting the quality standards of today and tomorrow and continuously ensures the safety and efficacy of nanoparticle-mediated therapies.
High-resolution analytical data is essential for the accurate and precise characterization of nanoparticles
Precisely analyzing Nanoparticle morphology at high resolution is often challenging and notoriously difficult to perform using established biophysical assays. Classical approaches that utilize light scattering, such as DLS or other light-optical techniques, are routinely used in attempt to characterize the size distribution of LNP formulations, but only work well on homogeneous samples with defined compositions and typically rely on hard to meet idealized assumptions of the underlying mathematical models. Furthermore, the resolution and analytical depth, even in the ideal examples, is typically too low to precisely characterize morphological changes in the nanometer (nm) range. Structural changes in the nanometer range, however, are often pivotal in determining the function of nanoparticle drugs, which typically exhibit a size range of 30-70 nm.
Nano-imaging techniques such as cryo-electron microscopy (cryo-EM) therefore offer the only real option to characterize LNP morphology at (sub-) nanometer resolution. However, as manual image analysis is typically cumbersome and reliable automated image annotation did not exist in the past, cryo-EM based imaging was typically solely used as a qualitative technique to merely visualize the exemplary morphology of a statistically insignificant sample of particles from a given LNP formulation.
Automating the analysis of Nano-Imaging Data is the key to obtaining statistically significant, accurate and precise results
ATEM has now overcome this limitation by developing an Artificial Intelligence (AI) based Neural Network to reliably identify and analyze all LNPs on any given number of cryo-EM images. By fully automating the image analysis procedure, it is now possible to automatically (!) extract highly relevant LNP morphological parameters like particle size, particle shape and particle morphology from a given set of hundreds or thousands of cryo-EM images (micrographs) in seconds. Analyzing a comparable dataset of cryo-EM micrographs by hand would take a well-trained human operator days or weeks. Hence, immediately rendering a manual analysis unfeasible.
By combining and analyzing the numerical data that is extracted from the morphology of every single LNP in the dataset, it is now possible to determine key particle quality attributes (like particle size, shape, and morphology) of any given LNP formulation at single particle precision and at scale. Automatically analyzing thousands to ten-thousands of single particles finally enables ATEM to generate statistically significant, highly accurate and precise summary reports that represent significant and unprecedentedly fine-granular statements on the quality of the analyzed LNP material.
Through its highly reliable and accurate LNP image annotation software, ATEM now moves morphological LNP characterization to the next, quantitative level.
Access available demo reports and quantification tools below to experience the analytical depth provided by the statistically significant analysis.
Making cryo-EM quantitative – demonstrating Statistical Significance, Accuracy and Precision
Extensive empirical validation studies show that the obtained quantitative results are statistically relevant, accurate and precise. Random draw studies from very large datasets (> 100.000 annotated LNP) show that at least 2,000 or 5,000 LNP should be analyzed to achieve, for example, 95% confidence intervals of < 1 nm in overall particle diameter for a given dataset.
Analyzing technical replicates from the same physical LNP sample shows that the repeatedly gathered results produce overall standard deviations of 1 nm or less for particle diameter, showing that the whole method is highly precise.
Finally, when evaluating accuracy in particle identification and classification by comparing human operator to Neural Network made selections on the same dataset, ATEM demonstrates that the AI powered particle annotation software generates human like results during automated particle annotation.
Consequently, automating the data generation and annotation processes not only critically improves the throughput of the analysis, but also guarantees the generation of statistically reliable data, which is essential for building trust with regulatory authorities and clinical confidence.
< 1.0 nm
95% Confidence Interval
(Avg. Particle Diameter, 5000 particles)
< 1.0 nm
Standard Deviation
(Avg. Particle Diameter, 5000 particles)
96.3%
Ø Classification Accuracy
(Particle Morphological Class)
Empowering LNP Development with Comprehensive Reporting
To make Customers benefit readily from the advanced technology, ATEM’s service model encompasses not only the analytical capabilities but also delivers detailed reporting that provides deep insights into the quality and characteristics of the analyzed LNP formulations, down to the single particle level, if required.
The all-inclusive service approach offered by ATEM, from sample preparation through cryo-EM data acquisition to detailed statistical analysis, enables Customers to quickly gain significantly deeper understandings of critical properties, stability, and quality parameters of their LNP formulations – immediately aiding in rationally optimizing formulation decisions and ensuring most stringent quality monitoring.
Takeaway
The now available highly detailed and statistically significant analysis of LNP morphology represents a significant leap forward in the development and quality control of nanoparticle-based drug delivery systems.
By combining cryo-EM nano-imaging with AI-driven analysis, ATEM offers an unprecedentedly detailed method to characterize LNPs at different stages of the development and manufacturing process. The new analytical insights not only increase trust in the quality of a (potential) drug product from regulatory authorities, but also offer the unique opportunity to formulation and development scientists to enhance the stability and predict the effectiveness novel nanoparticle-mediated drugs.
