Machine Learning Analysis Provides Insight into Mechanisms of Protein Particle Formation Inside Containers During Mechanical Agitation

Assessing subvisible particle risks in monoclonal antibodies: insights from quartz crystal microbalance with dissipation, machine learning, and in silico analysis

Throughout the lifecycle of biopharmaceutical development and manufacturing, monoclonal antibodies (mAbs) are subjected to diverse interfacial stresses and encounter various container surfaces. These interactions can cause the formation of subvisible particles (SVPs) that complicate developability and stability assessments of the drug products. This study leverages quartz crystal microbalance with dissipation (QCM-D), an interfacial characterization technique, as well as both in silico and experimentally measured physicochemical properties, to investigate the significant differences in SVP formation among different mAbs due to interfacial stresses. We conducted forced degradation experiments in borosilicate glass and high-density polyethylene containers, using agitation and stirring to rank 15 mAbs on SVP risks. Our data indicate that the kinetics of antibody adsorption to solid-liquid interfaces correlate strongly with SVP propensity in the stirring study yet show a weaker correlation with agitation-induced SVPs. In addition, SVP morphology was analyzed using self-supervised machine learning on flow imaging microscopy images. Despite the differing surface chemistry of the two container types, stirring resulted in similar SVP morphologies, in contrast to the unique morphologies produced by agitation. Collectively, our research demonstrates the utility of QCM-D and in silico models in evaluating mAb developability and their tendency to form interface-mediated SVPs, providing a strategy to mitigate risks associated with SVP formation in biotherapeutic development.

Assessment of biophysical properties of the first-in-class anti-cancer IgE antibody drug MOv18 IgE demonstrates monomeric purity and stability

Therapeutic monoclonal antibodies, which are almost exclusively IgG isotypes, show significant promise but are prone to poor solution stability, including aggregation and elevated solution viscosity at dose-relevant concentrations. Recombinant IgE antibodies are emerging cancer immunotherapies. The first-in-class MOv18 IgE, recognizing the cancer-associated antigen folate receptor-alpha (FRα), completed a Phase 1 clinical trial in patients with solid tumors, showing early signs of efficacy at a low dose. The inaugural process development and scaled manufacture of MOv18 IgE for clinical testing were undertaken with little baseline knowledge about the solution phase behavior of recombinant IgE at dose-relevant concentrations. We evaluated MOv18 IgE physical stability in response to environmental and formulation stresses encountered throughout shelf life. We analyzed changes in physical stability using multiple orthogonal analytical techniques, including particle tracking analysis, size exclusion chromatography, and multidetector flow field flow fractionation hyphenated with UV. We used dynamic and multiangle light scattering to profile aggregation status. Formulation at pH 6.5, selected for use in the Phase 1 trial, resulted in high monomeric purity and no submicron proteinaceous particulates. Formulation at pH 5.5 and 7.5 induced significant submicron and sub-visible particle formation. IgE formulation was resistant to aggregation in response to freeze-thaw stress, retaining high monomeric purity. Exposure to thermal stress at elevated temperatures resulted in loss of monomeric purity and aggregation. Agitation stress-induced submicron and subvisible aggregation, but monomeric purity was not significantly affected. MOv18 IgE retains monomeric purity in response to formulation and stress conditions, confirming stability. Our results offer crucial guidance for future IgE-based drug development.

Establishment of a subvisible particle profile in ophthalmic recombinant fusion protein and antibody formulations to control and monitor drug quality

The presence of subvisible particles in intravitreal injection ophthalmic formulations may affect drug efficacy and safety. Although Chapters 788 and 789 of the United States Pharmacopeia limit the concentration of particles ≥ 10 μm, particles less than 10 μm may cause blurred vision and floaters in the injected eyes. Due to limited access to ophthalmic protein formulations, few studies investigated the profile of subvisible particles in these formulations. In this study, the subvisible particle concentration, size distribution, and morphology of 11 intravitreal injection ophthalmic recombinant fusion protein and monoclonal antibody formulations at different drug clinical trial stages were characterized. In addition, there was significant difference in particle morphology, (circularity, compactness and etc) between unexpired and expired batches using Mann Whitney test, which was probably associated with the proportion change of protein and silicone oil particles and could be used for drug quality control at different clinical trial stages or on the market.

Running on the verge of collapse in lyophilization: What is the impact of the edge vial effect on colloidal protein stability and particle morphology?

Lyophilization is frequently employed to stabilize sensitive biopharmaceuticals. The potential importance of the edge vial effect, defined as the discrepancy in temperature and drying behavior between vials located in different positions of the batch, attracts considerable attention from the lyophilization community. Here, we lyophilized two model fusion-protein formulations by applying a design space representation to select conservative as well as aggressive lyophilization protocols on the verge of structural collapse of the cake. The effect of vial position and number of neighbor vials on cake appearance, cake volume, and residual moisture content was quantified. Moreover, the stability of the fusion-protein was assayed in relation to monomer content, size distribution of submicron and subvisible particles, as well as the presence of visible particles. Additionally, the images of micron-sized particles were analyzed by using a machine-learning based "fingerprinting" method to identify potential morphological differences induced by the edge vial effect. We concluded that the investigated quality attributes, including cake appearance, residual moisture, particle burden, and particle morphology, were not significantly affected by the edge vial effect. In contrast, the aggressive lyophilization cycle on the verge of collapse showed a minor, formulation-dependent impact on cake volume and particle count.

Application of one-class classification using deep learning technique improves the classification of subvisible particles

Capturing subvisible particles using flow imaging microscopy is useful for evaluating protein aggregates that may induce immunogenicity. Automated labeling is desirable to distinguish harmless components such as silicone oil (SO) from subvisible particles. The one-class classifier, which requires only target class data for model establishment, is suitable for machine learning and proposes a useful solution for distinguishing a subject with heterogeneous but stable distributions, such as SO. However, the effectiveness of the application of one-class classifiers to subvisible particles remains unclear. In this study, we investigated whether deep learning techniques can improve the performance on a variety of images. We prepared datasets using SO and two types of protein aggregates: immunoglobulin G-derived aggregates (AggIgG) and albumin-derived aggregates (AggAlb). The deep-learning technique improved the classification scores for both AggIgG and AggAlb. The classification scores for AggIgG were more satisfactory than those for AggAlb. Cluster analysis revealed that one-class classification using deep learning techniques achieved excellent effectiveness across almost all clusters in classifying AggIgG. Collectively, the deep learning technique remarkably improved the one-class classification of subvisible particles of AggIgG and AggAlb. Combined with deep learning, one-class classification can contribute to the evaluation of subvisible particles, particularly for AggIgG.

Representative training data sets are critical for accurate machine-learning classification of microscopy images of particles formed by lipase-catalyzed polysorbate hydrolysis

Polysorbate 20 (PS20) is commonly used as an excipient in therapeutic protein formulations. However, over the course of a therapeutic protein product's shelf life, minute amounts of co-purified host-cell lipases may cause slow hydrolysis of PS20, releasing fatty acids (FAs). These FAs may precipitate to form subvisible particles that can be detected and imaged by various techniques, e.g., flow imaging microscopy (FIM). Images of particles can then be classified using supervised convolutional neural networks (CNNs). However, CNNs should be trained on representative images of particles which, as we demonstrate in this work, may be challenging to obtain. Here, we tested several rapid techniques to create FA particles and examined whether CNNs trained on microscopy images of these rapidly formed particles could accurately classify images of particles that had been produced by kinetically slower lipase-catalyzed hydrolysis of PS20. CNNs trained on images of rapidly produced particles were less accurate in classifying images of FA particles that had been produced by enzymatic hydrolysis of PS20 than CNNs trained with images of particles generated by the same slow hydrolysis, highlighting the importance of using representative image data sets for training CNN classifiers.

High-Speed Imaging-Based Particle Attribute Analysis of Spray-Dried Amorphous Solid Dispersions Using a Convolution Neural Network

Spray drying is a well-established method for preparing amorphous solid dispersion (ASD) formulations to improve the oral bioavailability of poorly soluble drugs. In addition to the characterization of the amorphous phase, particle attributes of spray-dried intermediates (SDIs), including particle size, morphology, and microstructure, need to be carefully studied and controlled for optimizing drug product performance. Although recent developments in microscopy technology have enabled the analysis of morphological attributes for individual SDI particles, a high-throughput method is highly desirable. In this work, a fingerprinting method exploiting high-speed dynamic imaging, laser diffraction (LD), and a convolutional neural network (CNN) was developed to characterize and quantify size and morphological distributions of particles in batches of spray-dried ASDs. This imaging technology enables the generation of hundreds of thousands of single-particle images in a few minutes that are analyzed by both unsupervised and supervised CNN models. The unsupervised data mining analysis demonstrated that a batch of SDI is a mixture of diverse particle subpopulations with varying sizes and morphological attributes. Motivated by this observation, we developed a CNN model that enabled rapid computation of the volumetric composition of the distinct particle subpopulations in a SDI batch, thus generating a morphological fingerprint. We implemented this high-speed imaging-based particle attribute analysis method to investigate SDIs containing hypromellose acetate succinate as a model system. The CNN fingerprint results enabled quantification of the changes in the morphological distribution of SDI batches prepared with variations in the spray drying process parameters, and the results were in line with the LD and electron microscopy data. Our experiments and analysis demonstrate the robustness and throughput of this fingerprinting approach for quantifying particle size and morphological distributions of individual SDI batches, which can help guide spray drying process development and thereby enable the development of a drug product with more robust process and optimized performance.

Particle formation in response to different protein formulations and containers: Insights from machine learning analysis of particle images

Subvisible particle count is a biotherapeutics stability indicator widely used by pharmaceutical industries. A variety of stresses that biotherapeutics are exposed to during development can impact particle morphology. By classifying particle morphological differences, stresses that have been applied to monoclonal antibodies (mAbs) can be identified. This study aims to evaluate common biotherapeutic drug storage and shipment conditions that are known to impact protein aggregation. Two different studies were conducted to capture particle images using micro-flow imaging and to classify particles using a convolutional neural network. The first study evaluated particles produced in response to agitation, heat, and freeze-thaw stresses in one mAb formulated in five different formulations. The second study evaluated particles from two common drug containers, a high-density polyethylene bottle and a glass vial, in six mAbs exposed solely to agitation stress. An extension of this study was also conducted to evaluate the impact of sequential stress exposure compared to exposure to one stress alone, on particle morphology. Overall, the convolutional neural network was able to classify particles belonging to a particular formulation or container. These studies indicate that storage and shipping stresses can impact particle morphology according to formulation composition and mAb.

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.

Development of an ELISA-based device to quantify antibody adsorption directly on medical plastic surfaces

Monoclonal antibodies (mAbs) encounter numerous interfaces during manufacturing, storage, and administration. While protein adsorption at the solid/liquid interface has been widely explored on model surfaces, a key challenge remains - the detection of very small amounts of adsorbed mAb directly on real medical surfaces. This study introduces a novel ELISA-based device, ELIBAG, a new tool for measuring mAb adsorption on medical bags. The efficacy of this device was highlighted by successfully confirming the adsorption of an IgG1 on two medical bag types: a polypropylene IV administration bag and a low-density polyethylene pharmaceutical manufacturing bag. We also investigated IgG1 adsorption on plastic model surfaces, revealing a similar range of mAb bulk concentration for surface saturation on both model and bag surfaces. This innovative device, characterized by its high-throughput and rapid approach, paves the way for extensive investigations into therapeutic proteins, such as mAbs, adsorption on a variety of medical or pharmaceutical surfaces, diverse adsorption conditions, and the influence of excipients employed in mAb formulation, which could enhance the knowledge of mAb interactions with plastic surfaces throughout their lifecycle.

Stain-Free Approach to Determine and Monitor Cell Heath Using Supervised and Unsupervised Image-Based Deep Learning

Cell-based medicinal products (CBMPs) are a growing class of therapeutics that promise new treatments for complex and rare diseases. Given the inherent complexity of the whole human cells comprising CBMPs, there is a need for robust and fast analytical methods for characterization, process monitoring, and quality control (QC) testing during their manufacture. Existing techniques to evaluate and monitor cell quality typically constitute labor-intensive, expensive, and highly specific staining assays. In this work, we combine image-based deep learning with flow imaging microscopy (FIM) to predict cell health metrics using cellular morphology "fingerprints" extracted from images of unstained Jurkat cells (immortalized human T-lymphocyte cells). A supervised (i.e., algorithm trained with human-generated labels for images) fingerprinting algorithm, trained on images of unstained healthy and dead cells, provides a robust stain-free, non-invasive, and non-destructive method for determining cell viability. Results from the stain-free method are in good agreement with traditional stain-based cytometric viability measurements. Additionally, when trained with images of healthy cells, dead cells and cells undergoing chemically induced apoptosis, the supervised fingerprinting algorithm is able to distinguish between the three cell states, and the results are independent of specific treatments or signaling pathways. We then show that an unsupervised variational autoencoder (VAE) algorithm trained on the same images, but without human-generated labels, is able to distinguish between samples of healthy, dead and apoptotic cells along with cellular debris based on learned morphological features and without human input. With this, we demonstrate that VAEs are a powerful exploratory technique that can be used as a process monitoring analytical tool.

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.

Advancing algorithmic drug product development: Recommendations for machine learning approaches in drug formulation

Artificial intelligence is a rapidly expanding area of research, with the disruptive potential to transform traditional approaches in the pharmaceutical industry, from drug discovery and development to clinical practice. Machine learning, a subfield of artificial intelligence, has fundamentally transformed in silico modelling and has the capacity to streamline clinical translation. This paper reviews data-driven modelling methodologies with a focus on drug formulation development. Despite recent advances, there is limited modelling guidance specific to drug product development and a trend towards suboptimal modelling practices, resulting in models that may not give reliable predictions in practice. There is an overwhelming focus on benchtop experimental outcomes obtained for a specific modelling aim, leaving the capabilities of data scraping or the use of combined modelling approaches yet to be fully explored. Moreover, the preference for high accuracy can lead to a reliance on black box methods over interpretable models. This further limits the widespread adoption of machine learning as black boxes yield models that cannot be easily understood for the purposes of enhancing product performance. In this review, recommendations for conducting machine learning research for drug product development to ensure trustworthiness, transparency, and reliability of the models produced are presented. Finally, possible future directions on how research in this area might develop are discussed to aim for models that provide useful and robust guidance to formulators.

Solid-Solid Interfacial Contact of Tubing Walls Drives Therapeutic Protein Aggregation During Peristaltic Pumping

Peristaltic pumping during bioprocessing can cause therapeutic protein loss and aggregation during use. Due to the complexity of this apparatus, root-cause mechanisms behind protein loss have been long sought. We have developed new methodologies isolating various peristaltic pump mechanisms to determine their effect on monomer loss. Closed-loops of peristaltic tubing were used to investigate the effects of peristaltic pump parameters on temperature and monomer loss, whilst two mechanism isolation methodologies are used to isolate occlusion and lateral expansion-relaxation of peristaltic tubing. Heat generated during peristaltic pumping can cause heat-induced monomer loss and the extent of heat gain is dependent on pump speed and tubing type. Peristaltic pump speed was inversely related to the rate of monomer loss whereby reducing speed 2.0-fold increased loss rates by 2.0- to 5.0-fold. Occlusion is a parameter that describes the amount of tubing compression during pumping. Varying this to start the contacting of inner tubing walls is a threshold that caused an immediate 20-30% additional monomer loss and turbidity increase. During occlusion, expansion-relaxation of solid-liquid interfaces and solid-solid interface contact of tubing walls can occur simultaneously. Using two mechanisms isolation methods, the latter mechanism was found to be most destructive and a function of solid-solid contact area, where increasing the contact area 2.0-fold increased monomer loss by 1.6-fold. We establish that a form of solid-solid contact mechanism whereby the contact solid interfaces disrupt adsorbed protein films is the root-cause behind monomer loss and protein aggregation during peristaltic 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.

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.