Combining Machine Learning and Backgrounded Membrane Imaging: A Case Study in Comparing and Classifying Different Types of Biopharmaceutically Relevant Particles

Past, Current, and Future: Application of Image Analysis in Small Molecule Pharmaceutical Development

The often-perceived limitations of image analysis have for many years impeded the widespread application of such systems as first line characterisation tools. Image analysis has, however, undergone a notable resurgence in the pharmaceutical industry fuelled by developments system capabilities and the desire of scientists to characterize the morphological nature of their particles more adequately. The importance of particle shape as well as size is now widely acknowledged. With the increasing use of modelling and simulations, and ongoing developments though the integration of machine learning and artificial intelligence, the utility of image analysis is increasing significantly driven by the richness of the data obtained. Such datasets provide means to circumvent the requirement to rely on less informative descriptors and enable the move towards the use of whole distributions. Combining the improved particle size and shape measurement and description with advances in modelling and simulations is enabling improved means to elucidate the link between particle and bulk powder properties. In addition to improved capabilities to describe input materials, approaches to characterize single components within multicomponent systems are providing scientists means to understand how their material may change during manufacture thus providing a means to link the behaviour of final dosage forms with the particle properties at the point of action. The aim is to provide an overview of image analysis and update readers with innovations and capabilities to other methods in the small molecule arena. We will also describe the use of AI for the improved analysis using image analysis.

Features in Backgrounds of Microscopy Images Introduce Biases in Machine Learning Analyses

Subvisible particles may be encountered throughout the processing of therapeutic protein formulations. Flow imaging microscopy (FIM) and backgrounded membrane imaging (BMI) are techniques commonly used to record digital images of these particles, which may be analyzed to provide particle size distributions, concentrations, and identities. Although both techniques record digital images of particles within a sample, FIM analyzes particles suspended in flowing liquids, whereas BMI records images of dry particles after collection by filtration onto a membrane. This study compared the performance of convolutional neural networks (CNNs) in classifying images of subvisible particles recorded by both imaging techniques. Initially, CNNs trained on BMI images appeared to provide higher classification accuracies than those trained on FIM images. However, attribution analyses showed that classification predictions from CNNs trained on BMI images relied on features contributed by the membrane background, whereas predictions from CNNs trained on FIM features were based largely on features of the particles. Segmenting images to minimize the contributions from image backgrounds reduced the apparent accuracy of CNNs trained on BMI images but caused minimal reduction in the accuracy of CNNs trained on FIM images. Thus, the seemingly superior classification accuracy of CNNs trained on BMI images compared to FIM images was an artifact caused by subtle features in the backgrounds of BMI images. Our findings emphasize the importance of examining machine learning algorithms for image analysis with attribution methods to ensure the robustness of trained models and to mitigate potential influence of artifacts within training data sets.

Supervised and unsupervised machine learning approaches for monitoring subvisible particles within an aluminum-salt adjuvanted vaccine formulation

Suspensions of protein antigens adsorbed to aluminum-salt adjuvants are used in many vaccines and require mixing during vial filling operations to prevent sedimentation. However, the mixing of vaccine formulations may generate undesirable particles that are difficult to detect against the background of suspended adjuvant particles. We simulated the mixing of a suspension containing a protein antigen adsorbed to an aluminum-salt adjuvant using a recirculating peristaltic pump and used flow imaging microscopy to record images of particles within the pumped suspensions. Supervised convolutional neural networks (CNNs) were used to analyze the images and create "fingerprints" of particle morphology distributions, allowing detection of new particles generated during pumping. These results were compared to those obtained from an unsupervised machine learning algorithm relying on variational autoencoders (VAEs) that were also used to detect new particles generated during pumping. Analyses of images conducted by applying both supervised CNNs and VAEs found that rates of generation of new particles were higher in aluminum-salt adjuvant suspensions containing protein antigen than placebo suspensions containing only adjuvant. Finally, front-face fluorescence measurements of the vaccine suspensions indicated changes in solvent exposure of tryptophan residues in the protein that occurred concomitantly with new particle generation during pumping.

Interfacial Adsorption Controls Particle Formation in Antibody Formulations Subjected to Extensional Flows and Hydrodynamic Shear

During their manufacturing and delivery to patients, therapeutic proteins are commonly exposed to various interfaces and to hydrodynamic shear forces. Although adsorption of proteins to solid-liquid interfaces is known to foster formation of protein aggregates and particles, the impact of shear remains controversial, in part because of experimental challenges in separating the effects of shear from those caused by simultaneous exposure to interfaces. Extensional flows (occurring when solutions flow through sudden contractions) exert localized elongational forces that have been suspected to be damaging to proteins. In this work, we measured aggregation and particle formation in formulations of polyclonal and monoclonal antibodies subjected to extensional flow, high shear (105 s-1) and exposure to stainless-steel/water interfaces. Modification of the surface charge at the stainless steel/water interface changed protein adsorption characteristics without altering shear profiles, enabling shear and interfacial interactions to be separated. Even under conditions where antibodies were subjected to high hydrodynamic shear and extensional flow, production of subvisible particles could be inhibited by modifying the stainless-steel surface charge to minimize antibody adsorption. Digital images of particles recorded by flow imaging microscopy (FIM) and analyzed with machine learning algorithms were consistent with a particle formation mechanism by which antibodies adsorb and aggregate at the stainless-steel/water interface and subsequently form particles when shear displaces the interfacial aggregates, transporting them into the bulk solution. Topographical differences measured using atomic force microscopy (AFM) supported the proposed mechanism by showing reduced levels of protein adsorption on surface-charge-modified stainless-steel.

Micro-flow imaging multi-instrument evaluation for sub-visible particle detection

Sub-visible particles (SVPs) in pharmaceutical products are a critical quality attribute, and therefore should be monitored during development. Although light obscuration (LO) and microscopic particle count tests are the primary pharmacopeial methods used to quantify SVPs, flow imaging methods like Micro-Flow Imaging (MFI™) appear to overcome shortcomings of LO such as limited sensitivity concerning smaller translucent SVPs in the size range < 10 µm. Nowadays, MFI™ is routinely utilized during development of biologicals. Oftentimes multiple devices are distributed across several laboratories and departments. This poses challenges in data interpretation and consistency as well as in the use of multiple devices for one purpose. In this study, we systematically evaluated seven MFI™ instruments concerning their counting and size precision and accuracy, using an inter-comparable approach to mimic daily working routine. Therefore, we investigated three different types of particles (i) NIST certified counting standards, (ii) protein-coated particles, and (iii) stress-induced particles from a monoclonal antibody. We compared the results to alternative particle detection methods: LO and Backgrounded Membrane Imaging (BMI). Our results showed that the precision and accuracy of particle count and size, as well as the comparability of instruments, depended on the particle source and its material properties. The various MFI™ instruments investigated showed high precision (<15 %) and data generated on different instruments were of the same order of magnitude within pharmacopeial relevant size ranges for NIST certified counting standards. However, we found limitations in the upper and lower detection limits, contrary to the limits claimed by the manufacturer. In addition, proteinaceous and protein-containing particles showed statistically significant differences in particle counts, while the measured particle diameters of all sizes were quite consistent.

An Intra-Company Analysis of Inherent Particles in Biologicals Shapes the Protein Particle Mitigation Strategy Across Development Stages

To better understand protein aggregation and inherent particle formation in the biologics pipeline at Novartis, a cross-functional team collected and analyzed historical protein particle issues. Inherent particle occurrences from the past 10 years were systematically captured in a protein particle database. Where the root cause was identified, a number of product attributes (such as development stage, process step, or protein format) were trended. Several key themes were revealed: 1) there was a higher propensity for inherent particle formation with non-mAbs than with mAbs; 2) the majority of particles were detected following manufacturing at scale, and were not predicted by the small-scale studies; 3) most issues were related to visible particles, followed by subvisible particles; 4) 50% of the issues were manufacturing related. These learnings became the foundation of a particle mitigation strategy across development and technical transfer, and resulted in a set of preventive actions. Overall, this study provides further insight into a recognized industry challenge and hopes to inspire the biopharmaceutical industry to transparently share their experiences with inherent particles formation.

Approaches to expand the conventional toolbox for discovery and selection of antibodies with drug-like physicochemical properties

Antibody drugs should exhibit not only high-binding affinity for their target antigens but also favorable physicochemical drug-like properties. Such drug-like biophysical properties are essential for the successful development of antibody drug products. The traditional approaches used in antibody drug development require significant experimentation to produce, optimize, and characterize many candidates. Therefore, it is attractive to integrate new methods that can optimize the process of selecting antibodies with both desired target-binding and drug-like biophysical properties. Here, we summarize a selection of techniques that can complement the conventional toolbox used to de-risk antibody drug development. These techniques can be integrated at different stages of the antibody development process to reduce the frequency of physicochemical liabilities in antibody libraries during initial discovery and to co-optimize multiple antibody features during early-stage antibody engineering and affinity maturation. Moreover, we highlight biophysical and computational approaches that can be used to predict physical degradation pathways relevant for long-term storage and in-use stability to reduce the need for extensive experimentation.